Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Meniscal Repair
Abstract
Catherine Laible, MD Historically, treatment of meniscus tears consisted of complete
Drew A. Stein, MD meniscectomy. Over the past few decades, however, the long-term
morbidities of meniscal removal, namely the early development of
Daniel N. Kiridly
knee osteoarthritis, have become apparent. Thus, management of
meniscal tears has trended toward meniscal preservation. Recent
technological advances have made repairs of the meniscus easier
and stronger. In addition, adjunctive therapies used to enhance the
healing process have advanced greatly in the past few years.
Today, with increased understanding of the impact of meniscal loss
and the principles of meniscal repair and healing, meniscal
preservation is viewed as an increasingly realistic and important
goal in the management of meniscus tears.
Figure 1 Figure 2
Anterior
C D
B E
0 1 2 3 Medial Lateral 3 2 1 0
Figure 3 Table 1
Indications for Meniscal Repair
Repair Indications
Schematic illustration of the types of meniscal tears. Note that the bucket- Although meniscal preservation is
handle tear has a morphology similar to that of the longitudinal or vertical important, only certain types of tears
tear but involves more displacement of the tear edges. are amenable to repair. Factors that
contribute to good healing potential
and low failure rates have been well
varus or valgus stress to the knee studied. The relative indications of
Diagnosis while internally or externally rotat- meniscal repair are summarized in
ing the leg. The test is positive when Table 1.
The diagnosis of a meniscal tear is
a pop or a click is palpated at the The vascular supply of a meniscal
typically clinical. Symptoms include
joint line as the knee is slowly ex- tear is the most important intrinsic
joint line tenderness, mechanical
tended. The sensitivity for this test is factor in healing. Most meniscal re-
symptoms of catching or a locking
48% and 65% and the specificity is pairs are attempted on tears that are
sensation, clicking on moving the 94% and 86% for the medial and close to the vasculature supply, that
knee, and intra-articular effusion. lateral menisci, respectively. is, in the red-red or red-white zone.
The clinical evaluation should in- The Thessaly test was described by Prospective studies evaluating clini-
clude assessing for joint line tender- Karachalios et al10 in 2005. The pa- cal and arthroscopic assessments of
ness, range of motion testing, the tient stands on the affected knee and healing have found that tears within
Apley grind test, the McMurray test, flexes it to 20°, then internally and 2 mm of the meniscal vascular rim
and the Thessaly test. Joint line ten- externally rotates the knee and body. have the highest rates of healing fol-
derness has a reported sensitivity of A positive test produces either pain lowing repair.13,14 Conversely, those
71% and 78% and specificity of at the joint line or a locking or catch- that lie >4 mm from the rim have
87% and 90% for medial and lateral ing sensation. The Thessaly test was high rates of failure following re-
meniscal tears, respectively.10 For the found to have sensitivity of 89% and pair.13,14 However, some studies have
Apley grind test, the patient lies 92% and specificity of 97% and reported successful repair of tears
prone with the knee flexed to 90°. 96% for the medial and lateral me- that extend into the avascular zone
The examiner assesses for pain by nisci, respectively.10 of the meniscus, especially in
performing internal and external ro- MRI is typically used to confirm a younger patients.15,16
tation of the leg while applying axial clinical diagnosis. However, its added The length of a tear affects its sta-
load. This test has a sensitivity of value in diagnosis has been disputed; its bility. Tears measuring <1 cm in
41% for both medial and lateral usefulness is largely based on the qual- length are generally considered sta-
tears and a specificity of 93% and ity of the MRI. A prospective study ble, and repair is usually unneces-
86% for medial and lateral meniscal showed accuracy of 73.7% with MRI sary.13,15,17 Tears measuring >4 cm in
tears, respectively. diagnosis and accuracy of 80.7% on length are unstable to the point that
In the McMurray test, the patient clinical examination.11 MRI is not attempted repairs often fail; thus,
lies supine and the knee is flexed to sufficiently accurate to show whether tears of this size are rarely repaired,
90°. Next, the examiner applies a a tear is repairable.12 either.13,14
Figure 6 For medial meniscus repair, the more recent mechanical study by
medial incision is made anterior to Aros et al29 found that with high-
the medial head of the gastrocnemius strength suture material, load to fail-
muscle, thereby exposing the cap- ure is the same regardless of suture
sule. For lateral incisions, the dissec- orientation.
tion is made anterior to the lateral All-inside repair devices were de-
head of the gastrocnemius. Care is veloped to reduce surgical time, pre-
taken to avoid neurovascular struc- vent complications resulting from ex-
tures. A sterile spoon or a speculum ternal approaches, and allow access
may be used to retrieve sutures and to tears of the posterior horn. First-
visualize the capsule. Sutures must be generation all-inside techniques in-
tied with the knee in relative exten- volved the insertion of rigid arrow or
sion to prevent capture of the poste- screw implant devices made of ab-
rior capsule of the knee as it folds on sorbable polymers. However, it
Arthroscopic image of a meniscal flexion, thus limiting extension. quickly became apparent that the de-
repair performed with the inside-out The inside-out technique is still vices were prone to breaking30 and to
suturing technique with horizontal
mattress sutures in a left knee. commonly used, although it is very damaging articular cartilage;31 they
difficult technically to repair tears in were abandoned for second-generation
the posterior horns of the menisci headless screws and arrows, which
with this technique.24 Although it has protruded less. These improved rigid
Repair Techniques proved to be effective, this technique fixation devices are still used, although
has a significant learning curve and recent studies have shown them to have
Initially, repairs of meniscal tears typically requires the presence of a less mechanical strength than suture
were approached from the periphery surgical assistant. repairs.32 Järvelä et al33 recently
of the meniscus without arthroscopic In the outside-in technique, sutures showed that of 42 meniscal repairs
instrumentation; thus, only the most are passed through the meniscus performed using meniscal screws and
peripheral tears could be accessed. from the outside, thus avoiding the arrows, 11 failed clinically on
The inside-out suturing technique more extensive incisions and retrac- follow-up, and some exhibited artic-
was the first one used for arthro- tions involved in inside-out repairs. ular cartilage damage.
scopic repair of meniscal tears, and it As with inside-out repairs, however, The third-generation all-inside re-
is still considered to be the standard outside-in repairs are largely limited pair devices involve the insertion of
of care for meniscal repair. to anterior portions of the medial sutures and suture fixators. These
In general, meniscal repair begins and lateral menisci.25 devices have been shown to be clini-
with a complete arthroscopic assess- Prospective studies have indicated cally effective. Grant et al26 found a
ment of the knee and full evaluation success with both techniques. In a pooled failure rate of 14.6% among
of the tear. In patients who require meta-analysis of isolated meniscus three studies in their meta-analysis.
repair, the margins of the tear are dé- repairs, Grant et al26 found a com- A bovine mechanical study showed
brided, with or without rasping. At bined 17% incidence of repair failure third-generation all-inside devices to
that point, the surgeon must decide with the inside-out technique and an have the same or slightly less load to
on a repair technique: inside-out, average Lysholm score of 87.8 on failure than horizontal or vertical
outside-in, all-inside, or a combina- follow-up. In a follow-up study of 41 mattress sutures.29 Third-generation
tion of these. patients with menisci repaired using all-inside suturing systems remain a
With the inside-out technique, su- the outside-in technique, Abdelkafy viable option for meniscal repair.
tures are inserted into the meniscus et al27 found that 5 patients (12%) Fourth-generation repair devices
using a needle cannula under arthro- required subsequent partial menis- allow placement of sutures in the
scopic visualization (Figure 6). The cectomy, and 36 patients had a mean meniscus without the aid of an exter-
needles with suture attached are Lysholm score of 87.3 at a mean of nal incision or a suture fixator sys-
passed on either side of the tear 11.7 years. tem. These new devices are self-
through the meniscus, then out the Mechanical studies have histori- adjusting, with the anchor located
knee through the capsule. An inci- cally shown that vertical mattress su- behind the capsule and with a sliding
sion is made in the skin, and the su- tures provide stronger fixation than knot that can be tensioned appropri-
tures are tied down to the capsule. do horizontal sutures.28 However, a ately by the surgeon. In a mechanical
Figure 8 Figure 9
animal models to result in faster pair location and type.46 During this the importance of adhering to the re-
healing and better defect fill.45 Cell- period, the patient also must refrain habilitation protocol postoperatively.
based therapy involving the growth from participating in sports that in-
of autologous chondrocytes on an volve running or cutting.
implanted scaffold as well as the in- Some studies have shown that an References
jection of specific growth factors to accelerated rehabilitation program,
stimulate meniscal cells and augment including unrestricted weight bearing Evidence-based Medicine: Levels of
meniscal repair are being tested in and a return to sports activity as evidence are described in the table of
animal studies. No formal studies soon as it is tolerated, is no less ef- contents. In this article, reference 33
have been published to date regard- fective than standard, more conser- is a level I study. Reference 35 is a
ing outcomes of these techniques. vative, meniscal repair rehabilitation level II study. References 11, 14, 18,
programs in preventing repair fail- 28, 29, 37, 40, 43, and 48 are level
ures.50 However, currently, there is III studies. References 4, 10, 12, 15-
Rehabilitation insufficient corroborated evidence to 17, 20-27, 32, 34, 42, 44, 47, and 49
support the use of accelerated reha- are level IV studies. References 1-3,
The postoperative limitations of me- bilitation following meniscal repair. 5-9, 13, 30, 31, 36, 38, 39, 41, 45,
niscal repair are markedly greater 46, and 50 are level V expert opin-
than those of partial meniscectomy. ion.
Because it is impossible to know for Summary References printed in bold type are
certain before surgery whether a me- those published within the past 5
niscal tear is repairable, it is impor- Arthroscopic management of menis- years.
tant that patients be well-informed cal injury is the most commonly per-
1. Kim S, Bosque J, Meehan JP, Jamali A,
and prepared for a potentially exten- formed orthopaedic procedure in the Marder R: Increase in outpatient knee
sive rehabilitation protocol. United States annually. Many studies arthroscopy in the United States: A
have demonstrated the long-term se- comparison of National Surveys of
In the first 4 to 6 weeks following
Ambulatory Surgery, 1996 and 2006.
repair, the patient should avoid knee quelae of partial and complete men- J Bone Joint Surg Am 2011;93(11):994-
flexion during weight bearing. The iscectomy, and as a result, the use of 1000.
surgeon may choose to allow the pa- meniscus repair has increased. 2. Athanasiou KA, Sanchez-Adams J:
Structure-function relationships of the
tient to bear weight with the knee Factors shown to be predictive of a
knee meniscus, in: Synthesis Lectures on
immobilized in extension or to allow successful meniscus repair include an Tissue Engineering: Engineering the
knee flexion with restricted weight acute, unstable, longitudinal, periph- Knee Meniscus. Morgan & Claypool,
2009, pp 1-27.
bearing. The exact time period and eral tear in a young patient with a
specific restrictions vary based on re- concomitant ACL reconstruction. 3. Arnoczky SP, Warren RF:
Microvasculature of the human
pair location and type.46 Pressure on For tears that do not fit into this cat- meniscus. Am J Sports Med 1982;10(2):
the menisci is significantly increased egory but are deemed appropriate 90-95.
during flexion while weight bear- for surgical repair, multiple augmen- 4. Fairbank TJ: Knee joint changes after
ing,47 and the femoral condyles tation techniques are available to op- meniscectomy. J Bone Joint Surg Br
1948;30(4):664-670.
translate posteriorly relative to the timize healing potential.
5. Kawamura S, Lotito K, Rodeo SA:
tibia during this motion.48 Both of The inside-out technique remains Biomechanics and healing response of
those actions can put undue stress on the standard of care for meniscus re- the meniscus. Operative Techniques in
the repair in the early stages of heal- Sports Medicine 2003;11(2):68-76.
pair; however, the outside-in and the
ing. After this initial period, the pa- three most recent generations of all- 6. Ahmed AM, Burke DL: In-vitro
measurement of static pressure
tient should begin a physical therapy inside repairs all have been shown in distribution in synovial joints: Part I.
program that is tailored to the loca- the literature to be effective options Tibial surface of the knee. J Biomech
tion and type of repair performed.46 Eng 1983;105(3):216-225.
for tear fixation. Knowledge of the
Because of the increased strain on surrounding neurovascular struc- 7. Makris EA, Hadidi P, Athanasiou KA:
The knee meniscus: Structure-function,
the meniscus in tibial rotation and tures as well as the complications as- pathophysiology, current repair
deep flexion, deep squats and exces- sociated with each specific meniscal techniques, and prospects for
regeneration. Biomaterials 2011;32(30):
sive internal and external rotation of repair device is essential. Following
7411-7431.
the tibia should be avoided until at meniscus repair, a strict rehabilita-
8. Bedi A, Kelly NH, Baad M, et al:
least 4 months postoperatively.48,49 tion protocol is required, and the pa- Dynamic contact mechanics of the
The duration varies depending on re- tient must understand preoperatively medial meniscus as a function of radial
tear, repair, and partial meniscectomy. Arthroscopic repair of meniscus tears meniscal repairs: An in vitro
J Bone Joint Surg Am 2010;92(6):1398- extending into the avascular zone with biomechanical evaluation. Am J Sports
1408. or without anterior cruciate ligament Med 2010;38(9):1838-1844.
reconstruction in patients 40 years of age
9. Cooper DE, Arnoczky SP, Warren RF: and older. Arthroscopy 2000;16(8):822- 33. Järvelä S, Sihvonen R, Sirkeoja H,
Meniscal repair. Clin Sports Med 1991; 829. Järvelä T: All-inside meniscal repair with
10(3):529-548. bioabsorbable meniscal screws or with
21. Gallacher PD, Gilbert RE, Kanes G, bioabsorbable meniscus arrows: A
10. Karachalios T, Hantes M, Zibis AH, Roberts SN, Rees D: Outcome of
Zachos V, Karantanas AH, Malizos KN: prospective, randomized clinical study
meniscal repair prior compared with with 2-year results. Am J Sports Med
Diagnostic accuracy of a new clinical test concurrent ACL reconstruction. Knee
(the Thessaly test) for early detection of 2010;38(11):2211-2217.
2012;19(4):461-463.
meniscal tears. J Bone Joint Surg Am
34. Gunes T, Bostan B, Erdem M, Asci M,
2005;87(5):955-962. 22. Cipolla M, Scala A, Gianni E, Puddu G:
Sen C, Kelestemur MH: Biomechanical
Different patterns of meniscal tears in
11. Miller GK: A prospective study evaluation of arthroscopic all-inside
acute anterior cruciate ligament (ACL)
comparing the accuracy of the clinical ruptures and in chronic ACL-deficient meniscus repairs. Knee Surg Sports
diagnosis of meniscus tear with magnetic knees: Classification, staging and timing Traumatol Arthrosc 2009;17(11):1347-
resonance imaging and its effect on of treatment. Knee Surg Sports 1353.
clinical outcome. Arthroscopy 1996; Traumatol Arthrosc 1995;3(3):130-134.
12(4):406-413. 35. Kim JH, Chung JH, Lee DH, Lee YS,
23. Pujol N, Beaufils P: Healing results of Kim JR, Ryu KJ: Arthroscopic suture
12. Bernthal NM, Seeger LL, Motamedi K, meniscal tears left in situ during anterior anchor repair versus pullout suture
et al: Can the reparability of meniscal cruciate ligament reconstruction: A repair in posterior root tear of the medial
tears be predicted with magnetic review of clinical studies. Knee Surg meniscus: A prospective comparison
resonance imaging? Am J Sports Med Sports Traumatol Arthrosc 2009;17(4): study. Arthroscopy 2011;27(12):1644-
2011;39(3):506-510. 396-401. 1653.
13. Scott GA, Jolly BL, Henning CE: 24. Elkousy H, Higgins LD: Zone-specific 36. Aldrich D, Anschuetz R, LoPresti C,
Combined posterior incision and inside-out meniscal repair: Technical Fumich M, Pitluk H, O’Brien W:
arthroscopic intra-articular repair of the limitations of repair of posterior horns of Pseudoaneurysm complicating knee
meniscus: An examination of factors medial and lateral menisci. Am J Orthop arthroscopy. Arthroscopy 1995;11(2):
affecting healing. J Bone Joint Surg Am (Belle Mead NJ) 2005;34(1):29-34. 229-230.
1986;68(6):847-861.
25. Lambert EW, Bonner KF: Arthroscopic 37. Cohen SB, Boyd L, Miller MD: Vascular
14. Cannon WD Jr, Vittori JM: The meniscus repair with sutures: Outside-in. risk associated with meniscal repair
incidence of healing in arthroscopic Sports Medicine & Arthroscopy Review using Rapidloc versus Fas T-Fix:
meniscal repairs in anterior cruciate 2004;12(1):25-36. Comparison of two all-inside meniscal
ligament-reconstructed knees versus devices. J Knee Surg 2007;20(3):235-
stable knees. Am J Sports Med 1992; 26. Grant JA, Wilde J, Miller BS, Bedi A: 240.
20(2):176-181. Comparison of inside-out and all-inside
techniques for the repair of isolated 38. Jurist KA, Greene PW III, Shirkhoda A:
15. Noyes FR, Barber-Westin SD: meniscal tears: A systematic review. Am J Peroneal nerve dysfunction as a
Arthroscopic repair of meniscal tears Sports Med 2012;40(2):459-468. complication of lateral meniscus repair:
extending into the avascular zone in A case report and anatomic dissection.
patients younger than twenty years of 27. Abdelkafy A, Aigner N, Zada M, Arthroscopy 1989;5(2):141-147.
age. Am J Sports Med 2002;30(4):589- Elghoul Y, Abdelsadek H, Landsiedl F:
600. Two to nineteen years follow-up of 39. Chen NC, Martin SD, Gill TJ: Risk to
arthroscopic meniscal repair using the the lateral geniculate artery during
16. Rubman MH, Noyes FR, Barber-Westin outside-in technique: A retrospective arthroscopic lateral meniscal suture
SD: Arthroscopic repair of meniscal tears study. Arch Orthop Trauma Surg 2007; passage. Arthroscopy 2007;23(6):642-
that extend into the avascular zone: A 127(4):245-252. 646.
review of 198 single and complex tears.
Am J Sports Med 1998;26(1):87-95. 28. Rimmer MG, Nawana NS, Keene GC, 40. Choi NH, Kim TH, Victoroff BN:
Pearcy MJ: Failure strengths of different Comparison of arthroscopic medial
17. Tengrootenhuysen M, Meermans G, meniscal suturing techniques. meniscal suture repair techniques: Inside-
Pittoors K, van Riet R, Victor J: Long- Arthroscopy 1995;11(2):146-150. out versus all-inside repair. Am J Sports
term outcome after meniscal repair. Knee Med 2009;37(11):2144-2150.
Surg Sports Traumatol Arthrosc 2011; 29. Aros BC, Pedroza A, Vasileff WK, Litsky
19(2):236-241. AS, Flanigan DC: Mechanical 41. Henning CE, Lynch MA, Yearout KM,
comparison of meniscal repair devices Vequist SW, Stallbaumer RJ, Decker KA:
18. Matsubara H, Okazaki K, Izawa T, et al: with mattress suture devices in vitro. Arthroscopic meniscal repair using an
New suture method for radial tears of Knee Surg Sports Traumatol Arthrosc exogenous fibrin clot. Clin Orthop Relat
the meniscus: Biomechanical analysis of 2010;18(11):1594-1598. Res 1990;(252):64-72.
cross-suture and double horizontal
suture techniques using cyclic load 30. Calder SJ, Myers PT: Broken arrow: A 42. Ochi M, Uchio Y, Okuda K, Shu N,
testing. Am J Sports Med 2012;40(2): complication of meniscal repair. Yamaguchi H, Sakai Y: Expression of
414-418. Arthroscopy 1999;15(6):651-652. cytokines after meniscal rasping to
promote meniscal healing. Arthroscopy
19. Allaire R, Muriuki M, Gilbertson L, 31. Kumar A, Malhan K, Roberts SN: 2001;17(7):724-731.
Harner CD: Biomechanical consequences Chondral injury from bioabsorbable
of a tear of the posterior root of the screws after meniscal repair. Arthroscopy 43. Uchio Y, Ochi M, Adachi N, Kawasaki
medial meniscus: Similar to total 2001;17(8):34. K, Iwasa J: Results of rasping of
meniscectomy. J Bone Joint Surg Am meniscal tears with and without anterior
2008;90(9):1922-1931. 32. Brucker PU, Favre P, Puskas GJ, von cruciate ligament injury as evaluated by
Campe A, Meyer DC, Koch PP: Tensile second-look arthroscopy. Arthroscopy
20. Noyes FR, Barber-Westin SD: and shear loading stability of all-inside 2003;19(5):463-469.
44. Zhang Z, Arnold JA, Williams T, Sports Phys Ther 2012;42(3):274-290. 49. Tienen TG, Buma P, Scholten JG, van
McCann B: Repairs by trephination and Kampen A, Veth RP, Verdonschot N:
suturing of longitudinal injuries in the 47. Becker R, Wirz D, Wolf C, Göpfert B, Displacement of the medial meniscus
avascular area of the meniscus in goats. Nebelung W, Friederich N: Measure- within the passive motion characteristics
Am J Sports Med 1995;23(1):35-41. ment of meniscofemoral contact pressure of the human knee joint: An RSA study
after repair of bucket-handle tears with in human cadaver knees. Knee Surg
45. Strauss EJ, Hart JA, Miller MD, Altman biodegradable implants. Arch Orthop Sports Traumatol Arthrosc 2005;13(4):
RD, Rosen JE: Hyaluronic acid Trauma Surg 2005;125(4):254-260. 287-292.
viscosupplementation and osteoarthritis:
Current uses and future directions. Am J 48. Johal P, Williams A, Wragg P, Hunt D, 50. Barber FA: Accelerated rehabilitation for
Sports Med 2009;37(8):1636-1644. Gedroyc W: Tibio-femoral movement in meniscus repairs. Arthroscopy 1994;
the living knee: A study of weight 10(2):206-210.
46. Noyes FR, Heckmann TP, Barber-Westin bearing and non-weight bearing knee
SD: Meniscus repair and transplantation: kinematics using ‘interventional’ MRI.
A comprehensive update. J Orthop J Biomech 2005;38(2):269-276.
Abstract
Trevor Gaskill, MD Scapulothoracic bursitis and snapping scapula syndrome are rare
Peter J. Millett, MD, MSc diagnoses that contribute to considerable morbidity in some
patients. These conditions represent a spectrum of disorders
characterized by pain with or without mechanical crepitus. They are
commonly identified in young, active patients who perform
repetitive overhead activities. Causes include anatomic scapular or
thoracic variations, muscle abnormalities, and bony or soft-tissue
masses. Three-dimensional CT and MRI aid in detecting these
abnormalities. Nonsurgical therapy is the initial treatment of choice
but is less successful than surgical management in patients with
anatomic abnormalities. In many cases, scapular stabilization,
postural exercises, or injections eliminate symptoms. When
nonsurgical treatment fails, open and endoscopic techniques have
From the Naval Medical Center
been used with satisfactory results. Familiarity with the
Portsmouth, Portsmouth, VA
(Dr. Gaskill), and The Steadman neuroanatomic structures surrounding the scapula is critical to
Clinic, Vail, CO (Dr. Millett). avoid iatrogenic complications. Although reported outcomes of both
Dr. Millett or an immediate family open and endoscopic scapulothoracic decompression are
member has received royalties from
Arthrex; serves as a paid consultant
encouraging, satisfactory outcomes have not been universally
to Arthrex; has stock or stock achieved.
options held in Game Ready and
VuMedi; and has received research
or institutional support from Arthrex,
OrthoRehab, Össur Americas,
Siemens USA, Smith & Nephew,
and ConMed Linvatec. Neither
S capulothoracic articulation is
unique in that the concave scap-
ula must glide smoothly on the con-
drome has improved, and surgical
management has evolved. Patient-
reported outcome measures and return
Dr. Gaskill nor any immediate family
member has received anything of vex thoracic cage to provide a func- to sport have been integrated into out-
value from or has stock or stock tional foundation for glenohumeral come studies; assessment of results
options held in a commercial
motion. Although several bursal and based on these stringent criteria suggest
company or institution related
directly or indirectly to the subject of soft-tissue planes facilitate this mo- that although significant improvement
this article. tion, it remains an innately irregular of symptoms can be achieved with sur-
The views expressed in this article articulation. Despite this irregularity, gery, complete resolution is not univer-
are those of the author(s) and do bursitis and snapping scapula syn- sal. Therefore, despite recent advances,
not necessarily reflect the official drome remain relatively rare diagno- diagnosis and management of snapping
policy or position of the Department
ses. Symptoms vary from intermit- scapula syndrome remains challenging.
of the Navy, the Department of
Defense, or the US Government. tent, mildly symptomatic bursitis to
debilitating, recalcitrant crepitus.
J Am Acad Orthop Surg 2013;21: Anatomy and Physiology
214-224 Early recognition and management
are essential to minimize unnecessary
http://dx.doi.org/10.5435/ Bony Structure
JAAOS-21-04-214
disability.
Since its initial description in 1867 Scapulothoracic articulation is
Copyright 2013 by the American
by Boinet,1 understanding of the path- unique in that it does not exhibit a
Academy of Orthopaedic Surgeons.
omechanics of snapping scapula syn- true synovial articulation. Scapular
Scapulothoracic Bursae
Both anatomic and adventitial bur-
Illustration of right posterior shoulder demonstrating anatomic and adventitial
bursae of the scapulothoracic joint. sae within the scapulothoracic artic-
ulation have been described3 (Figure
1). Anatomic bursae are thought to
be physiologic; they facilitate gliding
of surfaces within the scapulotho-
motion and stability are controlled Muscular Anatomy racic joint (Figure 2). In anatomic
dynamically through muscular con- studies, two bursae have been consis-
Normal scapulothoracic motion is
traction. The scapula itself lies be- tently found: the infraserratus bursa
the result of several muscles working
tween the second and seventh ribs and the supraserratus bursa, which
in concert to orient the scapula be-
and is characterized by three borders are separated by the serratus anterior
cause no direct bony articulation ex-
(medial, lateral, and superior) and muscle.3 The infraserratus bursa fa-
ists. Dysfunction of any of these
three angles (inferomedial, supero- cilitates gliding of the serratus ante-
muscles may result in abnormal
medial, and lateral). The costal sur- rior muscle on the chest wall, while
scapulothoracic motion. The trape- the supraserratus bursae divides the
face of the scapula undulates, and
thickness ranges from 10.5 to 26.7 zius has a midline origin and inserts serratus anterior and subscapularis
mm.2 The superior and inferior scap- along the superior lip of the scapular muscles.4
ular angles exhibit considerable vari- spine, superficial to the rhomboid Adventitial bursae are typically
ability; the angle of junction with the musculature. It is innervated by the considered pathologic and are com-
scapular body ranges from 124° to spinal accessory nerve, which runs monly present near the superior or
162°.2 The ventral surface of the deep to the muscle with the trans- inferior scapular angles. Identifica-
scapula is concave and articulates on verse cervical artery. The concave tion of these bursae is inconsistent in
the convex chest wall. The supra- scapula glides along the convex chest anatomic studies, and their presence
scapular notch lies along the medial wall, using the serratus anterior and is thought to represent a pathologic
aspect of the lateral third of the su- subscapularis muscles as a soft-tissue state.5,6 Symptoms occurring along
perior border of the scapula. The articulation. The subscapularis origi- the inferior scapular angle are typi-
transverse scapular ligament sepa- nates from the anterior surface of the cally secondary to bursal tissue lying
rates the suprascapular nerve within scapula and is innervated along its between the serratus anterior and the
the notch from the suprascapular ar- anterior border by the upper and chest wall.7,8 Debate exists, however,
tery coursing above it. lower subscapular nerves. In con- regarding the anatomic location of
Neurovascular Anatomy
Several neurovascular structures
course within the scapulothoracic
articulation to innervate the sur-
rounding musculature (Figure 3). A
Illustration of an axial cross-section of the scapulothoracic articulation. Note thorough understanding of the neu-
the location of bursae in relation to muscular planes. roanatomy of this region is critical to
minimize the risk of iatrogenic com-
plications. Ultimately, the risk to in-
dividual neurovascular structures is
Figure 3 dependent upon the surgical ap-
proach used for bursal decompres-
sion. The spinal accessory nerve is
located along the middle section of
the levator scapulae muscle deep to
the trapezius muscle.11 The main
branches of the spinal accessory
nerve are at risk of injury with portal
placement cranial to the scapular
spine or with inadvertent dissection
during open approaches. Typically,
the dorsal scapular nerve is identified
deep to the rhomboid major and mi-
nor muscles, 1 to 2 cm medial to the
medial border of the scapula.11 The
transverse scapular artery gives rise
to deep and superficial branches,
with the deep branch forming the
dorsal scapular artery and accompa-
nying the dorsal scapular nerve. The
superficial branch runs along the spi-
nal accessory nerve. These structures
Illustration of the right posterior shoulder demonstrating the gross location of are at risk of injury with medial
neurovascular structures important in scapulothoracic articulation. Black dots placement of arthroscopic portals or
indicate typical portal locations, with the distance from medial scapular
border noted. medial dissection during an open sur-
gical approach.
The long thoracic nerve innervates namic disorders that are the result of ondary to either scapular or thoracic
the serratus anterior muscle and runs predisposing abnormal anatomy cage anatomic variation.
along the anterior belly of the mus- combined with sufficient scapulotho-
cle. Unless dissection is performed racic motion. Symptoms may be ex- Scapulothoracic Masses
considerably more lateral than is typ- perienced with minimal activity if
Several scapulothoracic masses can
ical, this nerve is infrequently at risk considerable bony abnormality is
also result in crepitation. Osteochon-
of injury during arthroscopic or open present. Alternatively, repetitive
droma of the rib or scapula is the
surgical techniques. After branching overuse may incite symptoms even in
most frequently reported mass re-
from the brachial plexus, the supra- the presence of normal anatomy and
sponsible for scapulothoracic symp-
scapular nerve and artery pass to- may result in a spectrum of severity
toms, and resection is frequently cu-
ward the suprascapular notch before ranging from mild soreness to debili-
innervating the rotator cuff muscula- tating crepitus. rative18,19 (Figure 4). Infrequently,
ture. The suprascapular nerve and Scapulothoracic bursitis is thought scapular chondrosarcoma may be en-
artery are at risk of injury if a su- to be the result of inflammation countered in older patients, thereby
peromedial scapular resection is per- caused by overuse of the shoulder emphasizing the need for diligent di-
formed or a superior arthroscopic girdle. If sufficient irritation is pres- agnostic evaluation. The presence of
portal is used. Placement of the por- ent to create a chronic inflammatory an elastofibroma dorsi in the in-
tal at the junction of the medial two environment, fibrosis of the affected frascapular region is also occasion-
thirds and lateral third of the supe- bursa may occur. Fibrosis may ulti- ally reported. In this location, the
rior scapular border generally pro- mately lead to recalcitrant bursitis or mass is capable of elevating the infe-
vides ≥10 mm of distance from the snapping even in the absence of overt rior scapular border, thereby creating
suprascapular nerve.12 The use of scapular masses or muscular abnor- mechanical symptoms.20
bony landmarks for superomedial malities.6,10
border resection generally aids the Overt scapular snapping is fre- Scapular Angulation
surgeon in maintaining a 2-to 3-cm quently the result of bony or soft- The angle of the superomedial scapula
margin from the suprascapular tissue masses within the scapulotho- may also predispose some patients to
notch.12,13 racic articulation and may coexist scapulothoracic dysfunction. The su-
with scapulothoracic bursitis. A peromedial scapula and scapular body
spectrum of mechanical causes re- normally form an angle that measures
Pathophysiology sults in crepitus, including fibrotic or between 124° and 162°, and anatomic
anomalous musculature7,14,15 and abnormalities in this area may result in
Under physiologic conditions, the malunion of scapular or rib frac- incongruity, focal loading, inflamma-
concave anterior scapula glides tures.16 Variability in the anatomy of tion and, if the angle is of sufficient
smoothly over the convex thoracic the costal or scapular surfaces is magnitude, crepitus.2 Cadaver stud-
cage during shoulder motion. This common and may also influence ies suggest that approximately 6% of
motion is critical to provide a stable scapulothoracic snapping.2 This is scapulae may demonstrate some de-
foundation for glenohumeral func- exemplified by scapulothoracic crep- gree of superomedial hooking, and
tion. The scapulothoracic articula- itus associated with scoliosis and ky- 8.6% of scapular specimens exhibit
tion is unique in that it glides upon phosis17 and the resolution of symp- superomedial angulation ≥35°.21
muscular layers rather than cartilagi- toms with postural training in some Similar bony abnormality is occa-
nous surfaces, and it has been de- patients.10 In addition, resection of sionally identified along the inferior
scribed as perhaps the most incon- the first rib for management of tho- scapular angle, which appears to be
gruent articulation in the human racic outlet syndrome may elicit the second most common site of
body. To this end, the presence of snapping of the scapula.15 In a study symptoms.21 Osteochondroma or a
bursal tissue is physiologic and nec- of 13 patients with thoracic outlet Luschka tubercle (ie, a bony protu-
essary to facilitate normal scapu- syndrome who underwent first-rib berance of the superomedial scapular
lothoracic motion. However, as in resection and later developed snap- border) can also result in persistent
other musculoskeletal disorders, ab- ping scapula syndrome, scapula symptoms.7 It is important to note,
normal stress persistently applied to symptoms resolved in 9 patients however, that scapular crepitus may
normal structures can result in dys- (70%) after resection of the supero- be present in asymptomatic patients.
function. Thus, scapulothoracic syn- medial scapular border. Therefore, Therefore, mechanical symptoms
dromes can be thought of as dy- scapular crepitation may occur sec- alone should not be considered
Figure 6
Illustrations demonstrating the technique for open scapular resection and bursectomy. A, An incision is made overlying
the medial scapular border. B, The trapezius is split and retracted, exposing the deep posterior shoulder musculature.
C, Muscles are detached subperiosteally as necessary to adequately expose the scapula for resection (dashed line).
D, After scapular resection, sutures are used to reattach the detached muscles to the scapular border.
eral to the superomedial scapular an- subperiosteally, with care taken to crepitation or recurrent bursitis if in-
gle and deep to the trapezius, is protect the dorsal scapular nerve lo- dicated. Ultimately, the detached
critical.34 To provide adequate scapu- cated 2 cm medially. After bursal ex- musculature is reattached through
lar exposure, the rhomboids and le- cision, the angle of the scapula can scapular drill holes with heavy non-
vator scapulae muscles are elevated be excised to alleviate mechanical absorbable suture. A similar tech-
scapulothoracic bursectomy. Signifi- cessful in providing adequate relief 10. Kuhn JE, Plancher KD, Hawkins RJ:
Symptomatic scapulothoracic crepitus
cant improvement of symptoms was of symptoms. Regardless of the sur- and bursitis. J Am Acad Orthop Surg
noted at a minimum 2-year follow- gical approach used, familiarity with 1998;6(5):267-273.
up. Moreover, the authors noted that the neuroanatomic structures that 11. Ruland LJ III, Ruland CM, Matthews
those patients who underwent bur- surround the scapula is critical to LS: Scapulothoracic anatomy for the
sectomy and partial scapulectomy arthroscopist. Arthroscopy 1995;11(1):
avoid iatrogenic complications. Al-
52-56.
experienced better outcomes than though reported outcomes of open
did those who underwent bursec- 12. Aggarwal A, Wahee P, Aggarwal AK,
and endoscopic scapulothoracic de- Kaur H, Sahni D: Anatomical
tomy alone.39 Pearse et al28 reported compression are encouraging, satis- considerations for safe scapular resection
similar results in a study of 13 pa- factory outcomes are not universally in snapping scapula syndrome. Surg
tients; 9 patients reported improve- Radiol Anat 2012;34(1):43-47.
achieved.
ment in symptoms following bursec- 13. Bell SN, van Riet RP: Safe zone for
tomy with or without partial arthroscopic resection of the
superomedial scapular border in the
scapulectomy. Thus, arthroscopic References treatment of snapping scapula syndrome.
management appears to be a reliable J Shoulder Elbow Surg 2008;17(4):647-
649.
management method for snapping Evidence-based Medicine: Levels of
scapula syndrome. 14. Milch H: Partial scapulectomy for
evidence are described in the table of snapping of the scapula. J Bone Joint
Surgical results are now commonly contents. In this article, references 1, Surg Am 1950;32(3):561-566.
measured based on patient-specific 3, 5-8, 14-16, 18-20, 22-30, 32, and 15. Wood VE, Verska JM: The snapping
outcome measures and the ability to 39-41 are level IV studies. Reference scapula in association with the thoracic
return to the prior level of sport, in outlet syndrome. Arch Surg 1989;
35 is level V expert opinion. 124(11):1335-1337.
addition to symptomatic improve-
References printed in bold type are 16. Takahara K, Uchiyama S, Nakagawa H,
ment. Recent literature suggests reli- Kamimura M, Ohashi M, Miyasaka T:
those published within the past 5
able improvement of symptoms can Snapping scapula syndrome due to
be achieved with open or arthro- years. malunion of rib fractures: A case report.
J Shoulder Elbow Surg 2004;13(1):
scopic techniques; however, complete 1. Bionet: Fait clinique. Bull Mem Soc Chir 95-98.
resolution of symptoms may not be Paris1867;8:458.
17. Manske RC, Reiman MP, Stovak ML:
achieved universally. Millett et al39 2. Aggarwal A, Wahee P, Harjeet, Nonoperative and operative
reported that although patients’ Aggarwal AK, Sahni D: Variable osseous management of snapping scapula. Am J
anatomy of costal surface of scapula and Sports Med 2004;32(6):1554-1565.
symptoms substantially improved its implications in relation to snapping
18. Fukunaga S, Futani H, Yoshiya S:
following arthroscopic procedures, scapula syndrome. Surg Radiol Anat
2011;33(2):135-140. Endoscopically assisted resection of a
outcome scores remained lower than scapular osteochondroma causing
expected. Similarly, Pearse et al28 3. Ciullo JV: Subscapular bursitis: snapping scapula syndrome. World J
Treatment of “snapping scapula” or Surg Oncol 2007;5:37.
noted that only six of nine patients “wash-board syndrome.” Arthroscopy
19. van Riet RP, Van Glabbeek F:
returned to their previous level of 1992;8:412-413.
Arthroscopic resection of a symptomatic
sport. 4. Kuhne M, Boniquit N, Ghodadra N, snapping subscapular osteochondroma.
Romeo AA, Provencher MT: The Acta Orthop Belg 2007;73(2):252-254.
snapping scapula: Diagnosis and
treatment. Arthroscopy 2009;25(11): 20. Cinar BM, Akpinar S, Derincek A, Beyaz
S, Uysal M: Elastofibroma dorsi: An
Summary 1298-1311.
unusual cause of shoulder pain
5. Cobey MC: The rolling scapula. Clin [Turkish]. Acta Orthop Traumatol Turc
Scapulothoracic bursitis and snap- Orthop Relat Res 1968;60:193-194. 2009;43(5):431-435.
ping scapula syndrome represent a 6. Percy EC, Birbrager D, Pitt MJ: 21. Edelson JG: Variations in the anatomy of
spectrum of disease with symptoms Snapping scapula: A review of the the scapula with reference to the
literature and presentation of 14 snapping scapula. Clin Orthop Relat Res
that range from mildly irritating to patients. Can J Surg 1988;31(4):248- 1996;(322):111-115.
debilitating. Nonsurgical therapy re- 250.
22. Arntz C, Matsen FI: Partial
mains the initial treatment of choice 7. Milch H: Snapping scapula. Clin Orthop scapulectomy for disabling
but appears to be less successful in 1961;20:139-150. scapulothoracic snapping. Orthop Trans
1990;14:252-253.
patients with anatomic scapulotho- 8. Sisto DJ, Jobe FW: The operative
treatment of scapulothoracic bursitis in 23. McCluskey G, Bigliani L: Surgical
racic abnormalities. If nonsurgical
professional pitchers. Am J Sports Med management of refractory
measures fail, surgical intervention, 1986;14(3):192-194. scapulothoracic bursitis. Orthop Trans
including open, arthroscopic, and 1991;15:801.
9. Codman E: The Shoulder. Malabar, FL,
combined techniques, have been suc- Krieger Publishing, 1984, pp 1-31. 24. Mozes G, Bickels J, Ovadia D, Dekel S:
The use of three-dimensional computed snapping scapula syndrome. J Shoulder scapulothoracic disorders, in Miller M,
tomography in evaluating snapping Elbow Surg 2002;11(1):80-85. Cole B, eds: Textbook of Arthroscopy.
scapula syndrome. Orthopedics 1999; Philadelphia, PA, Saunders, 2004,
22(11):1029-1033. 31. Kibler WB, McMullen J: Scapular
pp 277-287.
dyskinesis and its relation to shoulder
25. Higuchi T, Ogose A, Hotta T, et al: pain. J Am Acad Orthop Surg 2003; 37. Frank DK, Wenk E, Stern JC, Gottlieb
Clinical and imaging features of 11(2):142-151. RD, Moscatello AL: A cadaveric study of
distended scapulothoracic bursitis: the motor nerves to the levator scapulae
Spontaneously regressed pseudotumoral 32. Pavlik A, Ang K, Coghlan J, Bell S:
Arthroscopic treatment of painful muscle. Otolaryngol Head Neck Surg
lesion. J Comput Assist Tomogr 2004;
snapping of the scapula by using a new 1997;117(6):671-680.
28(2):223-228.
superior portal. Arthroscopy 2003;19(6):
38. Chan BK, Chakrabarti AJ, Bell SN: An
26. Ken O, Hatori M, Kokubun S: The MRI 608-612.
alternative portal for scapulothoracic
features and treatment of
scapulothoracic bursitis: Report of four 33. Groh GI, Simoni M, Allen T, Dwyer T, arthroscopy. J Shoulder Elbow Surg
cases. Ups J Med Sci 2004;109(1):57-64. Heckman MM, Rockwood CA Jr: 2002;11(3):235-238.
Treatment of snapping scapula with a
27. Hodler J, Gilula LA, Ditsios KT, periscapular muscle strengthening 39. Millett PJ, Gaskill TR, Horan MP, van
Yamaguchi K: Fluoroscopically guided program. J Shoulder Elbow Surg 1996; der Meijden OA: Technique and
scapulothoracic injections. AJR Am J 5(2):S6. outcomes of arthroscopic
Roentgenol 2003;181(5):1232-1234. scapulothoracic bursectomy and partial
34. Williams GR Jr, Shakil M, Klimkiewicz scapulectomy. Arthroscopy 2012;28(12):
28. Pearse EO, Bruguera J, Massoud SN, J, Iannotti JP: Anatomy of the 1776-1783.
Sforza G, Copeland SA, Levy O: scapulothoracic articulation. Clin
Arthroscopic management of the painful Orthop Relat Res 1999;(359):237-246. 40. Lien SB, Shen PH, Lee CH, Lin LC: The
snapping scapula. Arthroscopy 2006; effect of endoscopic bursectomy with
22(7):755-761. 35. Millett P, Pacheco I, Gobezie R, Warner mini-open partial scapulectomy on
JJ: Management of recalcitrant snapping scapula syndrome. J Surg Res
29. Lehtinen JT, Macy JC, Cassinelli E, scapulothoracic bursitis: Endoscopic 2008;150(2):236-242.
Warner JJ: The painful scapulothoracic scapulothoracic bursectomy and
articulation: Surgical management. Clin scapuloplasty. Techniques in Shoulder 41. Lehtinen JT, Tetreault P, Warner JJ:
Orthop Relat Res 2004;(423):99-105. and Elbow Surgery 2006;7:200-205. Arthroscopic management of painful and
stiff scapulothoracic articulation.
30. Nicholson GP, Duckworth MA: 36. O’Holleran J, Millett P, Warner JJ: Arthroscopy 2003;19(4):E28.
Scapulothoracic bursectomy for Arthroscopic management of
Chondroblastoma and
Chondromyxoid Fibroma
Abstract
Camila B. R. De Mattos, MD Chondroblastoma and chondromyxoid fibroma are benign but
Chanika Angsanuntsukh, MD locally aggressive bone tumors. Chondroblastoma, a destructive
lesion with a thin radiodense border, is usually seen in the
Alexandre Arkader, MD
epiphysis of long bones. Chondromyxoid fibroma presents as a
John P. Dormans, MD bigger, lucent, loculated lesion with a sharp sclerotic margin in the
metaphysis of long bones. Although uncommon, these tumors can
be challenging to manage. They share similarities in pathology that
could be related to their histogenic similarity. Very rarely,
From the Department of chondroblastoma may lead to lung metastases; however, the
Orthopaedic Surgery, The Children’s
Hospital of Philadelphia, mechanism is not well understood.
Philadelphia, PA (Dr. De Mattos and
Dr. Dormans), the Department of
Orthopaedic Surgery, Ramathibodi
Hospital, Mahidol University,
Bangkok, Thailand
(Dr. Angsanuntsukh), Children’s
C hondroblastoma is a rare, be-
nign bone tumor, usually lo-
cated in the epiphysis or apophysis
Epidemiology
Figure 1 mors, there is no single characteristic borders. The cells contain one or two
abnormality or chromosomal break- round, oval, slightly indented, or
ing point specific for chondroblas- even multilobulated nuclei with or
toma or CMF. without nucleoli.8 Occasional cells
The histogenesis of chondroblas- may have enlarged nuclei without
toma and CMF is still uncertain. Ro- nuclear atypia.4,25 The presence of
meo et al20 confirmed the active role mitotic figures is scarce.2,4,25 There
of cartilage-signaling molecules, both are scattered multinucleated osteo-
Indian Hedgehog/parathyroid hor- clast-type giant cells among the
mone–related protein (IHh/PTHrP) chondroblasts.7,8 There may be foci
and fibroblast growth factor, indicat- of chondroid matrix formed by the
ing that chondroblastoma is a neo- chondroblasts.25 Dystrophic calcifi-
Photomicrograph of
chondroblastoma. Note the diffuse plasm that originates from a mesen- cation is occasionally present and
and compact proliferation of chymal cell committed toward may surround individual cells, giving
mononuclear cells with indented chondrogenesis via active growth the classic “chicken wire” appear-
nuclei with abundant eosinophilic
plate signaling pathways. This con- ance7,8 (Figure 1), although this is
cytoplasm and distinct cell borders.
There is presence of focal clusion supports the chondrogenic not mandatory for diagnosis.
pericellular (ie, chicken wire) nature of this tumor and the close re- In 15% to 32% of cases, chondro-
calcification (arrows) on the top left lationship between the physis and blastoma may be associated with
corner (hematoxylin-eosin, original
magnification ×400). the tumor.21 CMF has myofibroblas- secondary aneurysmal bone cyst
tic differentiation in its “fibrous” ar- (ABC).7-9,11 Although the reasons for
eas driven by transforming growth this association are unclear, hypothe-
brae, generally in the posterior ele-
factor β-1.22 A strong expression of ses include mechanical stress,
ments and/or the body.14
the Sox9 gene, which is responsible trauma, and hemorrhage.13 More ag-
A prevalence of <0.5% of all bone tu-
for chondrocytic differentiation as gressive chondroblastoma that can
mors is reported in many series describ-
well as regulation of the expression cause metastases or recurrence shows
ing CMF.6,15,16 There is a slight male
of cartilage-specific genes in mature no difference in histology compared
predominance.17 CMF most com-
chondrocytes, especially the synthe- with less aggressive chondroblas-
monly arises in young patients in the
sis of collagen type II, was found in toma.25 The histology is equivalent
second or third decades of life.3,17,18
both chondroblastoma and CMF.23,24 to that of the primary site, and the
Most of these tumors are located in
This demonstrates that the expres- presence of atypical cells is rare.25
the metaphysis of long bones with
sion of Sox9 in these tumors is con- Grossly, CMF appears as lobulated,
variable distances from growth plate,
sistent with its commitment to the well-circumscribed, and sharply demar-
mainly in lower extremities.17 Rarely,
early phases of cartilage differentia- cated from the adjacent bone marrow.
the lesion involves the epiphysis. The
tion, with chondroblastoma being a The lesion is firm and white. The cut
diaphysis can be involved, especially
more “immature” tumor than CMF surface shows a solid tumor mass that
in large tumors. A lesion in small
because of the greater presence of is yellow, grayish-white, or blue-
bones, such as phalanges, may in-
positive Sox9 in CMF cells.23 gray.6 Microscopic analysis of CMF
volve the bone in its totality. The
reveals three components: myxoma-
proximal tibia is the most common
tous zones, fibrous zones, and fields
site, comprising 28% to 52% of all
Pathology that appear chondroid. The classic
lower extremity lesions in the litera-
histologic features of CMF are lob-
ture.17,19 This site is followed by the
Grossly, a chondroblastoma is a ules of stellate or spindle-shaped cells
ilium, ribs, distal femur, metatarsals,
gray-white tumor with yellowish ar- in abundant myxoid background or
and distal tibia.17 In contrast to
eas, usually because of calcification, chondroid intercellular material.
chondroblastoma, CMF is rarely en-
which can be soft, rubbery, or fria- Scattered giant cells are found in ap-
countered in the humerus.15,17
ble.2 Microscopically, chondroblas- proximately 50% of cases, usually at
toma reveals proliferation of mono- the edge of the lobules16,17 (Figure 2).
Etiology nuclear cells.10 The tumor is These lobules have a hypocellular
characterized by compact areas of center and a condensation of the nu-
Although cytogenetic abnormalities round, oval, or polygonal chondro- clei toward the periphery, creating a
can be highly specific for some tu- blasts with well-defined cytoplasmic hypercellular periphery. The inter-
Figure 2 Figure 3
Photomicrograph of chondromyxoid
fibroma. Note the stellate cells in a
myxoid background. The stellate
cells display mild atypia, but
mitoses are rarely seen. The lesion AP (A) and AP with internal rotation (B) radiographs of a chondroblastoma of
has a lobulated appearance, with the proximal humerus of a 15-year-old girl who presented with right shoulder
alternating cellular and mode pain lasting 4 months. The lesion is located entirely in the epiphysis, is
myxoid areas (hematoxylin-eosin, lucent, and has thin, sclerotic borders.
original magnification ×200).
Figure 4
Curettage and bone grafting of a chondroblastoma of the distal femur in an 11-year-old girl with chronic knee pain.
A, Preoperative AP radiograph demonstrating a lytic lesion on the distal femur that extends from the epiphysis into the
metaphysis. B, Sagittal T1-weighted magnetic resonance image demonstrating peripheral lobulation and associated
marrow edema. C, Intraoperative AP fluoroscopic image demonstrating curettage and bone grafting of the lesion
performed through a cortical window, thereby avoiding damage to the unaffected surrounding physis. D, AP radiograph
made at 7-month follow-up. The patient was asymptomatic, with no complications or recurrence.
At times, the lesion may be mottled from the epiphyseal lesion.21 A few scan shows increased uptake but sel-
or fuzzy or contain areas of calcifica- case reports of diaphyseal chondro- dom is needed for diagnosis.21
tion. Lesion size on radiograph var- blastoma have appeared in the litera- CMF classically presents as a lytic
ies, with most being <4 cm.2,5,8,11 Cal- ture.34 Chondroblastomas located on radiolucent medullary lesion with a
cifications are found especially in small bones may be more aggressive, thin sclerotic rim (Figure 6). In most
skeletally immature patients.7 with loss of cortical continuity and lesions, borders are sharp, with par-
It is uncommon to find periosteal bony destruction.35 tial or complete effacement of the
bone formation on radiographs, but CT can help define the anatomic cortex.16,17,29,36 CMF tends to be ec-
MRI usually depicts edema adjacent limits of the lesion, especially the dis- centrically located in the metaphysis
to the periosteum.31 In rare cases, es- tance to the growth plate and the re- of long bones. Rarely, or in advanced
pecially in older or neglected pa- lation of the lesion to subchondral cases, the lesion crosses the growth
tients, chondroblastoma may have bone. CT shows stipple calcification plate into epiphysis or extends into
an atypical presentation that clini- of the cartilaginous matrix, when the diaphysis.6 In small bones, CMF
cally and radiographically mimics an present. In addition, CT is useful in generally occupies the entire width of
aggressive process.5,28,32 delimiting lesions in unusual loca- the bone, causing thinning of cortices
The small percentage of chondro- tions as well as subchondral frac- and fusiform expansion of the bone.
blastomas with a secondary ABC in tures not visible on plain radio- The tumor typically has a scalloped
the histology usually show differ- graphs.13,14 border that is well defined by a nar-
ences from regular chondroblastoma In a few cases in which MRI was row rim of sclerotic bone. Chronic
on radiograph, and these sometimes not used in conjunction with radio- bone reaction and cortical thickening
lead to confusion in the diagnosis. graphs and the clinical presentation, are commonly present. Unusual peri-
Cystic changes are seen more com- this modality led to misdiagnosis or osteal reaction has been re-
monly when the lesion is located in overestimation of tumor aggressive- ported.16,17,36 Pseudotrabeculation,
bone, such as the patella.13,33 ness.31 Chondroblastoma usually is that is, ridges of the sclerotic rim at
Most chondroblastomas involve hypointense on T1-weighted images the edge of the lesion, is present.
the epiphysis of long bones.6 A small and variably ranges from hypoin- Gross and microscopic studies show
lesion is usually confined to a part of tense to hyperintense on T2- that there is no complete bony sep-
the epiphysis or apophysis, although weighted images, with or without tum.16
it may extend through the epiphyseal peripheral lobulation and the associ- Unlike other cartilaginous tumors,
plate21 (Figure 4). True metaphyseal ated marrow and soft-tissue edema calcification in CMF is unusual. The
chondroblastoma is rare; most re- that enhances after administration of prevalence of calcification in CMF
ported cases are of an extension contrast material31 (Figure 5). Bone ranges from 2.4% to 16% of cases
Differential Diagnosis
The differential diagnosis for chon-
droblastoma includes GCT, simple
bone cyst, ABC, enchondroma, eo-
sinophilic granuloma, fibrous dys-
plasia, clear cell chondrosarcoma,
subacute osteomyelitis (ie, Brodie ab-
scess), and, when a subchondral cyst
is present, Legg-Calvé-Perthes dis-
ease or osteochondritis dissecans.
Tuberculosis can mimic the periartic-
The same patient as in Figure 3. Postcontrast T1-weighted fat-suppressed
ular pain and bone lesion of chon-
coronal (A) and T2-weighted axial (B) magnetic resonance images
demonstrating heterogeneously increased signal. The bone marrow is droblastoma and should be consid-
normal. ered, especially in developing
countries.6,8,21,25
The differential diagnosis for CMF
Figure 6
includes benign lesions such as GCT,
simple bone cyst, ABC, enchondroma,
eosinophilic granuloma, fibrous dys-
plasia, osteoblastoma, osteofibrous
dysplasia, and nonossifying fibro-
mas.6,18 Malignant conditions that
must be differentiated are low-grade
chondrosarcoma and myxoid chon-
drosarcomas.
Management
The natural history of these tumors
is not completely understood; to
date, there has been no evidence of
potential spontaneous healing.7 Sur-
gical management is advised for both
types of tumors because no effective
medical management is available.
AP (A) and lateral (B) radiographs of the proximal tibia of a 22-year-old man Both chondroblastoma and CMF
with chondromyxoid fibroma who presented with mild pain of the left knee
lasting for 3 months. Note the mild expansion of the tumor and the scalloped generally have a favorable prognosis
borders defined by thin sclerotic bone. Biopsy was performed, and the when identified and managed appro-
diagnosis was confirmed. (Adapted and printed with permission from priately.
Dr. Olavo Pires de Carvalho, University of São Paulo, São Paulo, Brazil.)
The benchmark management of
chondroblastoma is curettage with
radiologically, and from 6.8% to MRI are the preferred imaging mo- bone grafting.8-12,25 The entire tumor
34% of cases histologically.16,36 Cal- dalities. CT demonstrates cortical in- should be excised, with the surgeon
cification presents more often in pa- tegrity and calcification of the matrix following meticulous oncologic crite-
tients aged >40 years and in flat well. MRI shows low signal on T1- ria of a thorough intralesional exci-
bones.36 Pathologic fractures may be weighted images and increased signal sion through a cortical and/or epi-
found but are unlikely.17 CT and on T2-weighted images and can help physeal window (Figure 4), avoiding
filled with bone graft. Of the 29 pa- those published within the past 5
tients with CMF in their study, there Summary years.
was no recurrence in the 3 who un-
Chondroblastoma and CMF are un- 1. Codman EA: The Classic: Epiphyseal
derwent curettage plus PMMA or chondromatous giant cell tumors of the
common benign bone tumors that upper end of the humerus. Surg Gynecol
the 4 who were treated with wide re-
present with insidious bone pain. Obstet.1931;52:543. Clin Orthop Relat
section. Res 2006;450:12-16.
Chondroblastoma usually involves
the epiphysis or apophysis of long 2. Jaffe HL, Lichtenstein L: Benign
chondroblastoma of bone: A
bones; CMF is a metaphyseal tumor. reinterpretation of the so-called
Metastasis The radiographic appearance of calcifying or chondromatous giant cell
chondroblastoma is of a lytic lesion tumor. Am J Pathol 1942;18(6):969-991.
Metastases from chondroblastoma
can arise from different primary with sclerotic borders. CMF is also 3. Jaffe HL, Lichtenstein L: Chondro-
myxoid fibroma of bone: A distinctive
sites. There is no reported relation of radiolucent with a sclerotic border, benign tumor likely to be mistaken espe-
metastasis to previous surgery or but it is usually larger than CMF and cially for chondrosarcoma. Arch Pathol
can have a bubbly appearance. Peri- (Chic) 1948;45(4):541-551.
nonsurgical treatment, tumor loca-
tion, or patient age.25,44,45 The inci- osteal reaction is uncommon for 4. Schajowicz F, Gallardo H: Epiphysial
chondroblastoma of bone: A clinico-
dence of metastases associated with both tumors. The chance of recur- pathological study of sixty-nine cases.
chondroblastoma is not known but rence of chondroblastoma is 8% to J Bone Joint Surg Br 1970;52(2):205-
13% and, for CMF, 20% to 25%. 226.
is thought to be very low. Rodgers
and Mankin46 described 2 patients of Surgical management can be chal- 5. Dahlin DC, Ivins JC: Benign
chondroblastoma: A study of 125 cases.
80 (2.5%) with chondroblastomas lenging because, especially in young
Cancer 1972;30(2):401-413.
treated for metastases at their institu- patients with chondroblastoma, the
6. Unni KK, Inwards CY: Chondromyxoid
tion. Selection bias can probably ex- ideal is to avoid the chance of recur- fibroma, in Unni KK, Inwards CY, eds:
plain this elevated rate; most authors rence while preserving the integrity Dahlin’s Bone Tumors, ed 6.
Philadelphia, PA, Wolters Kluwer,
believe it to be <1%.45 To date, there of the physis. Metastases are very
Lippincott Williams & Wilkins, 2010, pp
has been no published study on me- uncommon and have a good progno- 50-59.
tastases from CMF. sis when they are resectable. Addi- 7. Springfield DS, Capanna R, Gherlinzoni
tional genetic studies can likely help F, Picci P, Campanacci M: Chondro-
The lung is by far the most common blastoma: A review of seventy cases.
identify the cause of metastases and
site of distant metastases. Bones differ- J Bone Joint Surg Am 1985;67(5):748-
explain the nature of the aggressive 755.
ent from those of the primary site, soft
chondroblastoma. 8. de Silva MV, Reid R: Chondroblastoma:
tissue, the skin, and the liver are also
Varied histologic appearance, potential
cited.11,25,44 The time reported for me- diagnostic pitfalls, and clinicopathologic
tastases to manifest clinically ranges features associated with local recurrence.
Acknowledgments Ann Diagn Pathol 2003;7(4):205-213.
5 months to 33 years (average, 8
years) from the initial diagnosis of The authors would like to thank 9. Lin PP, Thenappan A, Deavers MT,
Lewis VO, Yasko AW: Treatment and
chondroblastoma.25,46 Olavo Pires de Carvalho, MD, and prognosis of chondroblastoma. Clin
Ostrowski et al47 reported evidence Marta E. Gutemberg, MD, for al- Orthop Relat Res 2005;438:103-109.
of p53 mutation in one patient with lowing the use of photographs of 10. Sailhan F, Chotel F, Parot R; SOFOP:
chondroblastoma and metastases. In their cases for this article. Chondroblastoma of bone in a pediatric
population. J Bone Joint Surg Am 2009;
contrast, Hasegawa et al27 found no 91(9):2159-2168.
evidence of p53 mutation in any of
11. Ramappa AJ, Lee FY, Tang P, Carlson
five patients with chondroblastoma References JR, Gebhardt MC, Mankin HJ:
without metastases. The p53 muta- Chondroblastoma of bone. J Bone Joint
Evidence-based Medicine: Levels of Surg Am 2000;82(8):1140-1145.
tion is a late event in tumorigenesis
and is present in many high-grade evidence are described in the table of 12. Suneja R, Grimer RJ, Belthur M, et al:
Chondroblastoma of bone: Long-term
sarcomas, including osteosarcomas contents. In this article, references results and functional outcome after
and chondrosarcomas.47 1-5, 7-12, 14-24, 26, 27, 30, 31, 33, intralesional curettage. J Bone Joint Surg
35-39, and 42 are level IV studies. Br 2005;87(7):974-978.
Patients usually survive several
years with metastatic lesions; the References 25, 29, 32, 34, 40, 41, 13. Fink BR, Temple HT, Chiricosta FM,
Mizel MS, Murphey MD: Chondro-
prognosis is better when the metasta- and 43-45 are level V expert opin- blastoma of the foot. Foot Ankle Int
ses are resectable.25,44 There is no re- ion. 1997;18(4):236-242.
ported benefit from chemotherapy.25 References printed in bold type are 14. Ilaslan H, Sundaram M, Unni KK:
Vertebral chondroblastoma. Skeletal Report of a fatal case with a review of calcification in chondromyxoid fibroma.
Radiol 2003;32(2):66-71. the literature on atypical, aggressive, and Skeletal Radiol 1998;27(10):559-564.
malignant chondroblastoma. Cancer
15. Dahlin DC: Chondromyxoid fibroma of 1985;55(8):1770-1789. 37. Lehner B, Witte D, Weiss S: Clinical and
bone, with emphasis on its morpho- radiological long-term results after
logical relationship to benign chondro- 26. Park HR, Park YK, Jang KT, Unni KK: operative treatment of chondroblastoma.
blastoma. Cancer 1956;9(1):195-203. Expression of collagen type II, S100B, Arch Orthop Trauma Surg 2011;131(1):
S100A2 and osteocalcin in chondro- 45-52.
16. Rahimi A, Beabout JW, Ivins JC, Dahlin blastoma and chondromyxoid fibroma.
DC: Chondromyxoid fibroma: A Oncol Rep 2002;9(5):1087-1091. 38. Copeland MM, Geschickter CF:
clinicopathologic study of 76 cases.
Chondroblastic tumors of bone: Benign
Cancer 1972;30(3):726-736. 27. Hasegawa T, Seki K, Yang P, et al:
and malignant. Ann Surg 1949;129(5):
Differentiation and proliferative activity
17. Wu CT, Inwards CY, O’Laughlin S, 724-733.
in benign and malignant cartilage tumors
Rock MG, Beabout JW, Unni KK: of bone. Hum Pathol 1995;26(8):838-
Chondromyxoid fibroma of bone: A 39. Thompson MS, Woodward JS Jr: The
845. use of the arthroscope as an adjunct in
clinicopathologic review of 278 cases.
Hum Pathol 1998;29(5):438-446. 28. Kirchhoff C, Buhmann S, Mussack T, the resection of a chondroblastoma of
et al: Aggressive scapular chondro- the femoral head. Arthroscopy 1995;
18. Lersundi A, Mankin HJ, Mourikis A, blastoma with secondary metastasis: A 11(1):106-111.
Hornicek FJ: Chondromyxoid fibroma: case report and review of literature. Eur
A rarely encountered and puzzling 40. Stricker SJ: Extraarticular endoscopic
J Med Res 2006;11(3):128-134.
tumor. Clin Orthop Relat Res 2005;439: excision of femoral head
171-175. 29. Sharma H, Jane MJ, Reid R: chondroblastoma. J Pediatr Orthop
Chondromyxoid fibroma of the foot and 1995;15(5):578-581.
19. Baker AC, Rezeanu L, O’Laughlin S, ankle: 40 years’ Scottish bone tumour
Unni K, Klein MJ, Siegal GP: registry experience. Int Orthop 2006; 41. Iwai T, Abe S, Miki Y, et al: A trapdoor
Juxtacortical chondromyxoid fibroma of 30(3):205-209. procedure for chondroblastoma of the
bone: A unique variant. A case study of femoral head: A case report. Arch
20 patients. Am J Surg Pathol 2007; 30. Armah HB, McGough RL, Goodman Orthop Trauma Surg 2008;128(8):763-
31(11):1662-1668. MA, et al: Chondromyxoid fibroma of 767.
rib with a novel chromosomal
20. Romeo S, Bovée JV, Jadnanansing NA, translocation: A report of four additional 42. Rybak LD, Rosenthal DI, Wittig JC:
Taminiau AH, Hogendoorn PC: cases at unusual sites. Diagn Pathol Chondroblastoma: Radiofrequency
Expression of cartilage growth plate 2007;2:44. ablation. Alternative to surgical resection
signalling molecules in chondroblastoma.
in selected cases. Radiology 2009;
J Pathol 2004;202(1):113-120. 31. Weatherall PT, Maale GE, Mendelsohn
DB, Sherry CS, Erdman WE, Pascoe HR: 251(2):599-604.
21. Maheshwari AV, Jelinek JS, Song AJ, Chondroblastoma: Classic and confusing
Nelson KJ, Murphey MD, Henshaw 43. Huvos AG, Marcove RC: Chondro-
appearance at MR imaging. Radiology blastoma of bone: A critical review. Clin
RM: Metaphyseal and diaphyseal 1994;190(2):467-474.
chondroblastomas. Skeletal Radiol Orthop Relat Res 1973;95:300-312.
2011;40(12):1563-1573. 32. Kim J, Kumar R, Raymond AK, Ayala
44. Khalili K, White LM, Kandel RA,
AG: Non-epiphyseal chondroblastoma
22. Romeo S, Eyden B, Prins FA, Briaire-de Wunder JS: Chondroblastoma with
arising in the iliac bone, and complicated
Bruijn IH, Taminiau AH, Hogendoorn by an aneurysmal bone cyst: A case multiple distant soft tissue metastases.
PC: TGF-beta1 drives partial report and review of the literature. Skeletal Radiol 1997;26(8):493-496.
myofibroblastic differentiation in Skeletal Radiol 2010;39(6):583-587.
chondromyxoid fibroma of bone. 45. Joshi DD, Anderson PM, Matsumoto J,
J Pathol 2006;208(1):26-34. 33. Strong DP, Grimer RJ, Carter SR, et al: Metastatic chondroblastoma with
Tillman RM, Abudu A: Chondro- elevated creatine kinase and
23. Dancer JY, Henry SP, Bondaruk J, et al: blastoma of the femoral head: Manage- paraneoplastic neurologic autoimmunity.
Expression of master regulatory genes ment and outcome. Int Orthop 2010; J Pediatr Hematol Oncol 2003;25(11):
controlling skeletal development in 34(3):413-417. 900-904.
benign cartilage and bone forming
tumors. Hum Pathol 2010;41(12):1788- 34. Azorín D, González-Mediero I, 46. Rodgers WB, Mankin HJ: Metastatic
1793. Colmenero I, De Prada I, López-Barea F: malignant chondroblastoma. Am J
Diaphyseal chondroblastoma in a long Orthop (Belle Mead NJ) 1996;25(12):
24. Konishi E, Nakashima Y, Iwasa Y, bone: First report. Skeletal Radiol 2006; 846-849.
Nakao R, Yanagisawa A: Immuno- 35(1):49-52.
histochemical analysis for Sox9 reveals 47. Ostrowski ML, Johnson ME, Truong
the cartilaginous character of chondro- 35. Tarkkanen M, Karaharju E, Böhling T, LD, Hicks MJ, Smith FE, Spjut HJ:
blastoma and chondromyxoid fibroma et al: Chromosome study of 249 patients Malignant chondroblastoma presenting
of the bone. Hum Pathol 2010;41(2): examined for a bone tumor. Clin Orthop as a recurrent pelvic tumor with DNA
208-213. Relat Res 1993;297:117-128. aneuploidy and p53 mutation as
supportive evidence of malignancy.
25. Kyriakos M, Land VJ, Penning HL, 36. Yamaguchi T, Dorfman HD: Skeletal Radiol 1999;28(11):644-650.
Parker SG: Metastatic chondroblastoma: Radiographic and histologic patterns of
Understanding Systematic
Reviews and Meta-analyses in
Orthopaedics
Abstract
Kelly A. Lefaivre, MD, MSc, The systematic literature review is a powerful tool for summarizing
FRCSC and evaluating current knowledge related to a specific research
Gerard P. Slobogean, MD, MPH, question. Systematic reviews have many advantages over
FRCSC traditional narrative reviews. A meta-analysis of data from a
systematic review can provide a better estimate of a treatment
effect than can individual studies. To ensure quality conclusions,
rigorous methods must be applied to systematic reviews, such as
formulation of a specific research question, systematic literature
search, selection and assessment of included studies, data
extraction, quality assessment of included studies, meta-analysis
and presentation of results, and formation of conclusions. Threats
to the internal validity and generalizability of the conclusions of
systematic reviews include lack of clarity or appropriateness of the
research question, poor quality of the included studies,
heterogeneity of results between studies, inappropriate
conclusions, and inappropriate application in clinical practice.
Table 2
Use of the PICOS Format for Systematic Literature Review
Population Intervention Comparator Outcomes Study Design Question
Patients undergo- Single-bundle Double-bundle 1. KT-1000 All comparative Using all comparative study
ing ACL recon- reconstruction reconstruction arthrometer study designs designs, is double-bundle su-
struction (MEDmetric) perior to single-bundle recon-
2. Pivot-shift test- struction when comparing
ing postoperative KT-1000 ar-
thrometer (MEDmetric) and
pivot-shift testing?
Patients undergo- No patellar resur- Patellar resurfac- 1. Frequency of RCTs Based on RCT data, does pa-
ing primary TKA facing ing reoperation tellar resurfacing decrease
2. Postoperative the incidence of reoperations
anterior knee and postoperative anterior
pain knee pain and improve knee
3. Improvement in scores compared with not
knee scores resurfacing the patella during
primary TKA?
Children with dis- Crossed K-wire Lateral-only Iatrogenic ulnar All comparative Using all comparative study
placed supracon- fixation K-wire fixation nerve injury study designs designs, what is the risk of
dylar fractures of iatrogenic ulnar nerve injury
the humerus with crossed K-wire fixation
versus lateral-only K-wire fix-
ation in children with dis-
placed supracondylar fracture
of the humerus?
ACL = anterior cruciate ligament; K-wire = Kirschner wire; PICOS = population, intervention, comparators, outcomes, study design;
RCT = randomized controlled trial; TKA = total knee arthroplasty
Figure 1
Hypothetical flow diagram of the study selection process for systematic literature reviews, with associated research
steps.
final stage and the reason for their Data extraction typically includes geneity in the measurement of out-
exclusion (Figure 1). general information about the arti- comes poses a significant challenge
cle, study characteristics, participant in the pooling of results of treatment.
Data Extraction characteristics, intervention, and The method of measuring outcomes
The fourth stage in every systematic outcome data. Outcome data are of- in orthopaedics has itself been the
review consists of data extraction ten the most complicated to extract topic of many systematic reviews.20,30
from the selected studies. The exact because attention needs to be paid
data points extracted should be not only to the outcome itself but Quality Assessment
clearly stated in the methods section also to how it was measured and Before attempting any synthesis of
and should be easy for the reader to how the measurement was standard- data, the authors must consider and
follow through to the results section. ized. In orthopaedic studies, hetero- report on the study quality. In broad
terms, a quality assessment answers Meta-analysis and of differences in study populations, in-
the question of how close to the Presentation of Results terventions, outcome measurement, or
“truth” are a study’s findings or con- study methodology. The Q, H, and I2
The systematic, reproducible sum-
clusions.12,28 It is important to detect are some of the statistical tests that
mary and presentation of studies an-
significant sources of bias because have been described to quantify het-
swering a specific research question
these can have as much influence on erogeneity in this setting. The I2 sta-
is in itself a systematic review (ie,
the outcome as any other factor. tistic is the easiest to interpret and
qualitative systematic review). In
Historically, quality assessment has the most commonly used in health-
consisted of the application of a list some cases, the review process is care meta-analyses. It provides a pro-
of criteria, such as those published in stopped and the findings presented at portion of the between-study vari-
the Cochrane Handbook for System- this stage because of study heteroge- ability in the intervention effect that
atic Reviews of Interventions, to neity, lack of appropriate studies, or is attributable to study heterogeneity,
evaluate the methodology of ran- other reasons. The combination of or how much of the variability is not
domized controlled trials.8 These cri- results from such studies is known as due to chance. It is generally ac-
teria include randomization and a meta-analysis (ie, quantitative sys- cepted that ≤25% is indicative of lit-
blinding, as well as evaluation for tematic review). tle heterogeneity, 50% shows moder-
confounders, study dropouts, and A meta-analysis is a statistical ate heterogeneity, and 75% is
intention-to-treat analysis. combination of results from two or indicative of a high degree of hetero-
Evaluation of bias in systematic re- more similar studies. This is not a geneity.33
views of observational, or lower- simple arithmetic combination of re- Results from systematic reviews
quality, studies is even more complex. sults; rather, it is a weighted average and meta-analyses typically have two
In these cases, the authors must con- that gives proportional weight to the key elements: a table of included
sider the appropriateness of the study number of subjects in each study. study characteristics and forest plots.
design, the sources of bias, choice of Further, the statistical methods of The table is a simple representation
outcome measure, statistical issues, meta-analysis take into account of the studies selected for inclusion
quality of reporting, intervention, and intra- and interstudy variability and typically identifies the author,
generalizability of the results.9 Sev- when combining the results. year of publication, source journal,
eral scales have been devised to rate The two types of model that are and other information relevant to the
the quality of observational studies. commonly used to statistically com- specifics of the systematic review.
The Newcastle-Ottawa Scale uses a bine results are the fixed effects The forest plot is a graphic represen-
star system in which each study is model and the random effects model. tation of the meta-analysis, with a
judged on three broad categories: se- The first assumes a constant treat- representation of the statistical
lection of the study groups, compa- ment effect between studies, whereas weight of each study as well as the
rability of the groups, and ascertain- the second assumes that the effect is CIs. An example of a forest plot is
ment of either the exposure or randomly distributed between the seen in Figure 2.34
outcome of interest for case-control studies. In short, the fixed effects Regression may be appropriate in
or cohort studies, respectively.31 model assumes that the study itself is meta-analyses in which other impor-
The findings of the quality assess- the only variable, whereas the ran- tant reported variables may have an
ment can be used by authors in sev- dom effects model introduces a sec- effect on the treatment effect. Meta-
eral ways. First, a narrative synthesis ond source of variation between regression is used in meta-analyses
of quality can be performed to ex- studies. Typically, the random effects to evaluate the effect of moderator
plain differences across results in model produces more conservative variables on a study effect using
studies included in the systematic re- estimates, indicated by wider confi- regression-based techniques (ie, esti-
view. Conversely, for a systematic re- dence intervals (CIs).32 mating the effect of the moderator
view that is carried on to a meta- Evaluation of study heterogeneity is variable while controlling for other
analysis, the findings can be another key element of the statistical variables in the model). Fixed-effect
incorporated into the data synthesis testing needed for a meta-analysis. In meta-regression does not allow for
using subgroup or sensitivity analy- this context, heterogeneity refers to the the specification of within- and
ses. Regardless of approach, study amount of variability between the re- between-study variation. Instead, the
quality should be clearly addressed sults of individual studies compared authors can use random-effects
in some way by the authors of a with what would be expected by meta-regression, accounting for
well-performed systematic review.12,32 chance. Variability could be the result between-study variation.35
Figure 2
Example of a forest plot of simple decompression versus transposition of the ulnar nerve for cubital tunnel syndrome.
The references cited in this figure do not appear in this article, and the data have not been verified for accuracy: Biggs
et al, Neurosurg, 2006; Bartels et al, Neurosurg, 2005; Taha et al, Neurosurg, 2004; Bimmler et al, Ann Hand Surgery,
1996; Davies et al, Aust NZ J Surg, 1991; Adelaar et al, J Hand Surg Am, 1984; Foster et al, J Hand Surg, 1981;
Chan et al, Neurosurg, 1980; MacNicol et al, J Bone Joint Surg Br, 1979; Paine et al, Can J Surg, 1970.
CI = confidence interval. (Reproduced with permission from Macadam SA, Gandhi R, Bezuhly M, Lefaivre KA: Simple
decompression versus anterior subcutaneous and submuscular transposition of the ulnar nerve for cubital tunnel
syndrome: A meta-analysis. J Hand Surg Am 2008;33[8]:1314.e1-12.)
cently published a meta-analysis that for between-study heterogeneity is an are generalizable to his or her clinical
followed these basic principles to en- important final step to determine scenario. This includes ensuring sim-
sure its validity and clinical useful- whether and how the results of mul- ilar patient populations, interven-
ness. In their review, they sought to tiple studies should be pooled. tions, and outcomes of interest. Con-
compare the outcomes of surgical These principles of data analysis sider the following clinical scenario.
versus nonsurgical care for displaced and methodology were applied to the A 14-year-old elite hockey player
midshaft clavicle fractures. A search meta-analysis of 16 observational presents with an acute midshaft clav-
for studies related to clavicle fracture studies in a systematic review of the icle fracture. The injury occurred
management would identify several effect of early surgery on hip fracture during the off-season, and it is com-
hundred articles; however, by defin- mortality.21 The methodologic qual- pletely displaced, with 2 cm of short-
ing their research question, the sys- ity of each included study was as- ening. The patient and family want
tematic review was limited to the sessed using the Newcastle-Ottawa to know whether surgical treatment
handful of relevant articles that com- Scale, and the authors tested clini- is the best option to restore shoulder
pare these two treatments in the cally plausible factors that could ex- function for the upcoming season.
specified midshaft fracture pattern. plain heterogeneity between studies. You counsel the family that the re-
Furthermore, to minimize the poten- More importantly, they recognized cent meta-analysis of randomized
tial of pooling studies with poor in- that differences in mortality between controlled trials by McKee et al38
ternal validity (ie, methodologic early and late treatment groups demonstrates better functional out-
flaws), the authors chose to include might have been due to confounding comes (weighted Constant score 94.3
only randomized controlled trials factors present in the observational versus 90.2) and lower nonunion
that compared surgical and nonsur- study designs. For example, sicker rates (RR 0.18; 95% CI, 0.06 to
gical treatments. Finally, the defined patients often require more preoper- 0.51) with surgical fixation than
outcomes of interest—patient func- ative evaluations than their healthier nonsurgical treatment. Is this advice
tion and treatment complications— counterparts. As a result, the authors appropriate?
ensured a balanced reporting of evi- performed stratified analyses, with A closer look at the trials included
dence that would not inadvertently one analysis pooling all included in the meta-analysis by McKee et al38
focus on the positive or negative studies and a second consisting of a shows that the clinical scenario and
characteristics of a given treatment. meta-analysis of the five studies that the meta-analysis populations are
Even when the methodology is cor- had been statistically adjusted for different. The mean patient age
rectly designed, several pitfalls re- age, sex, and comorbidities. Finally, among the included studies ranged
main in the data analysis stage. Even the authors provided additional from 25 to 41.3 years. Furthermore,
in a review that includes only ran- analyses of mortality at 30 days, 6 the current trial literature may not
domized controlled trials, method- months, and 1 year to improve the apply to adolescents because of the
ologic weakness within the included relevance of the results. This proved potential in children for correction of
studies can significantly influence the to be important to understanding the malalignment and shortening via late
conclusions of the review. As a re- treatment effects because, although closure of the clavicular physes.39
sult, the authors and readers must in- the benefits of early surgery are not Surgical fixation ultimately may be
terpret the results of any pooled significant at 30 days (relative risk the optimal treatment in the clinical
analysis in the context of the quality [RR] 0.90; 95% CI, 0.71 to 1.13), scenario presented; however, it is im-
of each individual study. Some au- substantial risk reduction is demon- portant to recognize that it is easy to
thors may choose to exclude low- strated at 1 year (RR 0.55; 95% CI, inappropriately apply the results of a
quality studies or to perform sensi- 0.40 to 0.75). This example demon- systematic review because important
tivity analyses to determine the strates that the conclusions of a methodologic details from each in-
influence of these studies on the meta-analysis can be misleading if cluded study may be omitted from
pooled results. In the data analysis clinically relevant stratification and the meta-analysis manuscript.
phase, it is important to ensure that contextualization are not performed.
the included studies are appropriate The final difficulty in using system-
to pool based on clinically similar atic review or meta-analysis data lies Critically Evaluating
patient populations, confounding in applying the results to clinical Systematic Reviews
factors, and other differences be- practice. The reader must critically
tween studies that may have influ- determine whether the individual Critical appraisal of a systematic re-
enced the results. Statistical testing studies pooled in the meta-analysis view or meta-analysis requires a
the quality of the included studies, studies in orthopaedic research. J Bone acute orthopaedic trauma. Injury 2010;
Joint Surg Am 2009;91(suppl 3):87-94. 41(8):787-803.
the heterogeneity between studies,
and the appropriateness of the pre- 10. Cook DJ, Mulrow CD, Haynes RB: Sys- 23. Guyatt GH, Sackett DL, Sinclair JC,
tematic reviews: Synthesis of best evi- et al: Users’ guides to the medical litera-
sentation of results and associated dence for clinical decisions. Ann Intern ture: IX. A method for grading health
conclusions. Med 1997;126(5):376-380. care recommendations. JAMA 1995;
274(22):1800-1804.
11. Bangert-Drowns RL: Misunderstanding
meta-analysis. Eval Health Prof 1995; 24. Khan KS, Kunz R, Kleijnen J, Antes G:
Systematic Reviews to Support Evidence-
References 18(3):304-314.
based Medicine: How to Review and
12. Glasziou P, Irwig L, Bain C, Colditz G: Apply Findings of Healthcare Research.
Evidence-based Medicine: Levels of Systematic Reviews in Health Care: A London, United Kingdom, Royal Society
Practical Guide. Cambridge, United of Medicine Press, 2003, pp 21-35.
evidence are described in the table of Kingdom, Cambridge University Press,
contents. In this article, references 17 2001, pp 1-108. 25. Suarez-Almazor ME, Belseck E, Homik
J, Dorgan M, Ramos-Remus C: Identify-
and 38 are level I studies. Reference 13. Meredick RB, Vance KJ, Appleby D, ing clinical trials in the medical literature
39 is level V expert opinion. Lubowitz JH: Outcome of single-bundle with electronic databases: MEDLINE
versus double-bundle reconstruction of alone is not enough. Control Clin Trials
References printed in bold type are the anterior cruciate ligament: A meta- 2000;21(5):476-487.
those published within the past 5 analysis. Am J Sports Med 2008;36(7):
1414-1421. 26. Slobogean GP, Verma A, Giustini D,
years. Slobogean BL, Mulpuri K: MEDLINE,
14. Pakos EE, Ntzani EE, Trikalinos TA: EMBASE, and Cochrane index most pri-
1. Evidence-Based Medicine Working Patellar resurfacing in total knee arthro- mary studies but not abstracts included
Group: Evidence-based medicine: A new plasty: A meta-analysis. J Bone Joint in orthopedic meta-analyses. J Clin
approach to teaching the practice of Surg Am 2005;87(7):1438-1445. Epidemiol 2009;62(12):1261-1267.
medicine. JAMA 1992;268(17):2420-
2425. 15. Slobogean BL, Jackman H, Tennant S, 27. Slobogean GP, Lefaivre KA, Nicolaou S,
Slobogean GP, Mulpuri K: Iatrogenic O’Brien PJ: A systematic review of
2. Guyatt G, Rennie D, Meade MO, Cook ulnar nerve injury after the surgical treat- thromboprophylaxis for pelvic and ac-
DJ, eds: User’s Guide to the Medical Lit- ment of displaced supracondylar frac- etabular fractures. J Orthop Trauma
erature: Essentials of Evidence-Based tures of the humerus: Number needed to 2009;23(5):379-384.
Clinical Practice, ed 2. Chicago, IL, harm, a systematic review. J Pediatr
AMA Press, and New York, NY, Orthop 2010;30(5):430-436. 28. Torgerson C: Systematic Reviews. Lon-
McGraw-Hill Professional, 2008, pp don, United Kingdom, Continuum Pub-
5-40. 16. Wright RW, Brand RA, Dunn W, Spin- lishing, 2003, pp 52-88.
dler KP: How to write a systematic re-
3. Bhandari M, Morshed S, Tornetta P III, view. Clin Orthop Relat Res 2007;455: 29. Edwards P, Clarke M, DiGuiseppi C,
Schemitsch EH: Design, conduct, and 23-29. Pratap S, Roberts I, Wentz R: Identifica-
interpretation of nonrandomized ortho- tion of randomized controlled trials in
paedic studies: A practical approach. 17. Bhandari M, Devereaux PJ, Swiont- systematic reviews: Accuracy and reli-
(All) evidence matters. J Bone Joint Surg kowski MF, et al: Internal fixation com- ability of screening records. Stat Med
Am 2009;91(suppl 3):1. pared with arthroplasty for displaced 2002;21(11):1635-1640.
fractures of the femoral neck: A meta-
4. Wilson P, Petticrew M, Calnan M, Naza- analysis. J Bone Joint Surg Am 2003; 30. Corrales LA, Morshed S, Bhandari M,
reth I; Medical Research Council’s Popu- 85(9):1673-1681. Miclau T III: Variability in the assess-
lation Health Sciences Research Network ment of fracture-healing in orthopaedic
Knowledge Transfer Project Team: Why 18. Sproul RC, Iyengar JJ, Devcic Z, Feeley trauma studies. J Bone Joint Surg Am
promote the findings of single research BT: A systematic review of locking plate 2008;90(9):1862-1868.
studies? BMJ 2008;336(7646):722. fixation of proximal humerus fractures.
Injury 2011;42(4):408-413. 31. Wells GA, Shea B, O’Connell D, et al:
5. Thacker SB: Meta-analysis: A quantita- The Newcastle–Ottawa Scale (NOS) for
tive approach to research integration. 19. Cheng SY, Levy AR, Lefaivre KA, Guy assessing the quality of nonrandomised
JAMA 1988;259(11):1685-1689. P, Kuramoto L, Sobolev B: Geographic studies in meta-analyses. Ottawa, On-
trends in incidence of hip fractures: A tario, Canada, Ottawa Health Research
6. Petticrew M: Why certain systematic re- comprehensive literature review. Institute. Available at: http://www.ohri.
views reach uncertain conclusions. BMJ Osteoporos Int 2011;22(10):2575-2586. ca/programs/clinical_epidemiology/
2003;326(7392):756-758. oxford.asp. Accessed January 17, 2013.
20. Lefaivre KA, Slobogean G, Starr AJ, Guy
7. Brown P, Brunnhuber K, Chalkidou K, P, O’Brien PJ, Macadam SA: Methodol- 32. Egger M, Davey Smith G, Altman D:
et al: How to formulate research recom- ogy and interpretation of radiographic Systematic Reviews in Health Care:
mendations. BMJ 2006;333(7572):804- outcomes in surgically treated pelvic Meta-Analysis in Context. London,
806. fractures: A systematic review. J Orthop United Kingdom, BMJ Publishing
Trauma 2012;26(8):474-481. Group, 2001, pp 122-210.
8. Higgins JP, Green S, eds: Cochrane
Handbook for Systematic Reviews of 21. Simunovic N, Devereaux PJ, Sprague S, 33. Higgins JP, Thompson SG: Quantifying
Interventions, version 5.1.0 (updated et al: Effect of early surgery after hip heterogeneity in a meta-analysis. Stat
March 2011). Oxfordshire, United King- fracture on mortality and complications: Med 2002;21(11):1539-1558.
dom, The Cochrane Collaboration, Systematic review and meta-analysis.
2011. CMAJ 2010;182(15):1609-1616. 34. Macadam SA, Gandhi R, Bezuhly M,
Lefaivre KA: Simple decompression ver-
9. Simunovic N, Sprague S, Bhandari M: 22. Clay FJ, Newstead SV, McClure RJ: A sus anterior subcutaneous and submus-
Methodological issues in systematic re- systematic review of early prognostic cular transposition of the ulnar nerve for
views and meta-analyses of observational factors for return to work following cubital tunnel syndrome: A meta-
analysis. J Hand Surg Am 2008;33(8): 38. McKee RC, Whelan DB, Schemitsch EH, JAMA 2000;283(15):2008-2012.
1314.e1-12. McKee MD: Operative versus nonopera-
tive care of displaced midshaft clavicular 41. Moher D, Liberati A, Tetzlaff J, Altman
35. Thompson SG, Higgins JP: How should fractures: A meta-analysis of randomized DG; PRISMA Group: Preferred reporting
meta-regression analyses be undertaken clinical trials. J Bone Joint Surg Am items for systematic reviews and meta-
and interpreted? Stat Med 2002;21(11): 2012;94(8):675-684. analyses: The PRISMA statement. BMJ
1559-1573. 2009;339:b2535.
39. Smekal V, Oberladstaetter J, Struve P,
36. GRADE working group website. http:// Krappinger D: Shaft fractures of the 42. Centre for Evidence-based Medicine
www.gradeworkinggroup.org. Accessed clavicle: Current concepts. Arch Orthop (CEBM): Critical appraisal. Available at:
January 17, 2013. Trauma Surg 2009;129(6):807-815. http://www.cebm.net/index.aspx?o=
1157. Accessed January 17, 2013.
37. Guyatt GH, Oxman AD, Schünemann 40. Stroup DF, Berlin JA, Morton SC, et al:
HJ, Tugwell P, Knottnerus A: GRADE Meta-analysis of observational studies in 43. Yuan Y, Hunt RH: Systematic reviews:
guidelines: A new series of articles in the epidemiology: A proposal for reporting. The good, the bad, and the ugly. Am J
Journal of Clinical Epidemiology. J Clin Meta-analysis Of Observational Studies Gastroenterol 2009;104(5):1086-1092.
Epidemiol 2011;64(4):380-382. in Epidemiology (MOOSE) group.
targeting the modulatory actions of NF-κB activity by annexins require Expression of early and late
differentiation markers (proliferating cell
annexins on NF-κB signaling may their ability to bind Ca2+. Therefore, nuclear antigen, syndecan-3, annexin VI,
provide a novel and selective way to interfering with the Ca2+-binding and alkaline phosphatase) by human
inhibit NF-κB signaling in diseases, ability of these annexins may emerge osteoarthritic chondrocytes. Am J Pathol
2001;159(5):1777-1783.
such as cancer and OA. as a potential novel strategy to spe-
Future research needs to determine cifically inhibit NF-κB activity in OA 4. Marcu KB, Otero M, Olivotto E, Borzi
RM, Goldring MB: NF-kappaB
strategies of how to interfere with and other diseases. signaling: Multiple angles to target OA.
the modulatory role of annexins on Curr Drug Targets 2010;11(5):599-613.
NF-κB signaling. Recent findings 5. Bist P, Leow SC, Phua QH, et al:
showing that AnxA4 modulates References Annexin-1 interacts with NEMO and
NF-κB signaling by directly interact- RIP1 to constitutively activate IKK
complex and NF-κB: Implication in
ing with the p50 unit of the heterodi- 1. Buckwalter JA, Saltzman C, Brown T: breast cancer metastasis. Oncogene
meric p50/p65 NF-κB complex in a The impact of osteoarthritis:
2011;30(28):3174-3185.
Implications for research. Clin Orthop
Ca2+-dependent manner6 suggest Relat Res 2004;(427 suppl):S6-S15. 6. Jeon YJ, Kim DH, Jung H, et al:
that the interactions of annexins 2. Fatimathas L, Moss SE: Annexins as
Annexin A4 interacts with the NF-
with NF-κB signaling components disease modifiers. Histol Histopathol kappaB p50 subunit and modulates NF-
2010;25(4):527-532. kappaB transcriptional activity in a
and, ultimately, the stimulation of Ca2+-dependent manner. Cell Mol Life
3. Pfander D, Swoboda B, Kirsch T: Sci 2010;67(13):2271-2281.
Figure 1 bone, tendon, and muscle regenera- eta T: Comparison of human stem cells
derived from various mesenchymal
tion. In a rabbit bone defect model, tissues: Superiority of synovium as a cell
Matsusaki et al12 demonstrated that source. Arthritis Rheum 2005;52(8):
using tissue-engineered construct de- 2521-2529.
rived from synovial-MSCs with hy- 5. Koga H, Muneta T, Ju YJ, et al: Synovial
droxyapatite accelerates osteoinduc- stem cells are regionally specified
according to local microenvironments
tion. In a rat Achilles tendon graft after implantation for cartilage regenera-
model, synovial-MSC implantation tion. Stem Cells 2007;25(3):689-696.
into bone tunnel accelerated early re- 6. Nakamura T, Sekiya I, Muneta T, et al:
modeling of tendon-to-bone heal- Arthroscopic, histological and MRI
ing.13 Another rat model study analyses of cartilage repair after a
minimally invasive method of
showed that synovial-MSCs have transplantation of allogeneic synovial
myogenic potential and contribute to mesenchymal stromal cells into cartilage
Histologic appearance of human defects in pigs. Cytotherapy 2012;14(3):
skeletal muscle regeneration in
synovial mesenchymal stem cells 327-338.
under the phase contrast vivo.14 In spite of these reports, there
microscope 14 days after initial is not sufficient evidence regarding 7. Bilge O, Doral MN, Atesok K, et al: The
effects of the synovium on chondrocyte
plating. osteogenic and myogenic potential of growth: An experimental study. Knee
synovial-MSCs compared with bone Surg Sports Traumatol Arthrosc 2011;
19(7):1214-1223.
that are in direct contact with the sy- marrow– and muscle-derived MSCs.
8. Horie M, Driscoll MD, Sampson HW,
novium produce more chondrocytes et al: Implantation of allogenic synovial
compared with cartilage grafts that Future Perspectives stem cells promotes meniscal
regeneration in a rabbit meniscal defect
are not. Studies of animal meniscal In vitro and animal model studies model. J Bone Joint Surg Am 2012;
defect models demonstrated that support the use of synovial-MSCs in 94(8):701-712.
transplanted synovial-MSCs adhere cartilage regeneration as an impor- 9. Horie M, Sekiya I, Muneta T, et al:
to sites of meniscal injury, differenti- tant, arguably superior, cell-based Intra-articular injected synovial stem
cells differentiate into meniscal cells
ate into cells resembling meniscal fi- treatment alternative. Further inves- directly and promote meniscal
brochondrocytes, and enhance me- tigations in the near future should regeneration without mobilization to
distant organs in rat massive meniscal
niscal regeneration.8,9 help in our understanding the com-
defect. Stem Cells 2009;27(4):878-887.
Sporadic human studies have re- plexity of synovial-MSC biology in
10. Morito T, Muneta T, Hara K, et al:
ported that the number of synovial- terms of isolation, characterization, Synovial fluid-derived mesenchymal stem
MSCs in synovial fluid increases in culturing, distinguishing from, and cells increase after intra-articular
knees with degenerated cartilage and interacting with other cell types be- ligament injury in humans.
Rheumatology (Oxford) 2008;47(8):
osteoarthritis and following intra- fore this promising cell-based ther- 1137-1143.
articular ligament injury.10,11 This ob- apy can be translated into clinical 11. Sekiya I, Ojima M, Suzuki S, et al:
servation raises the question whether practice. Human mesenchymal stem cells in
the number of synovial-MSCs that synovial fluid increase in the knee with
degenerated cartilage and osteoarthritis.
are mobilized from synovium into J Orthop Res 2012;30(6):943-949.
synovial fluid increases according to References
12. Matsusaki M, Kadowaki K, Tateishi K,
the degree of cartilage degeneration et al: Scaffold-free tissue-engineered
as part of the reparative process. Hu- 1. Fox DB, Warnock JJ: Cell-based construct-hydroxyapatite composites
meniscal tissue engineering: A case for generated by an alternate soaking
man trials investigating the effects of synoviocytes. Clin Orthop Relat Res process: Potential for repair of bone
intra-articular synovial-MSC trans- 2011;469(10):2806-2816. defects. Tissue Eng Part A 2009;15(1):
55-63.
plantation to promote cartilage re- 2. De Bari C, Dell’Accio F, Tylzanowski P,
generation and/or to prevent osteoar- Luyten FP: Multipotent mesenchymal 13. Ju YJ, Muneta T, Yoshimura H, Koga H,
stem cells from adult human synovial Sekiya I: Synovial mesenchymal stem
thritis should answer this question. membrane. Arthritis Rheum 2001;44(8): cells accelerate early remodeling of
1928-1942. tendon-bone healing. Cell Tissue Res
2008;332(3):469-478.
Application of Synovial- 3. Jones BA, Pei M: Synovium-derived stem
MSCs for Bone, Tendon, cells: A tissue-specific stem cell for 14. De Bari C, Dell’Accio F, Vandenabeele F,
cartilage engineering and regeneration. Vermeesch JR, Raymackers JM, Luyten
and Muscle Regeneration Tissue Eng Part B Rev 2012;18(4):301- FP: Skeletal muscle repair by adult
311. human mesenchymal stem cells from
Synovial-MSCs may offer an alterna- synovial membrane. J Cell Biol 2003;
tive cell-based treatment strategy for 4. Sakaguchi Y, Sekiya I, Yagishita K, Mun- 160(6):909-918.