ANZ J. Surg.

2002; 72: 153155

CASE REPORT

SPONTANEOUS SPLENIC RUPTURE SECONDARY TO METASTATIC GASTRIC CARCINOMA: CASE REPORT AND REVIEW
PHILIP SMART, MARK CULLINAN AND GARY CROSTHWAITE
Department of Surgery, Royal Melbourne Hospital, Parkville, Victoria, Australia Key words: gastric carcinoma, splenic metastases, spontaneous splenic rupture. Abbreviations: CT, computed tomography; DVT, deep vein thrombosis.

INTRODUCTION
Splenic metastases are a clinical rather than pathological rarity. The spleen is involved in metastatic malignancy in 613% of cases at autopsy study.1,2 Although traditionally thought to imply widespread metastatic disease, the advent of increasingly sophisticated imaging techniques used in the follow up of neoplastic patients has recently challenged this notion.3 Often asymptomatic, the clinical presentation of splenic metastases may be unusual and range from splenomegaly to left upper quadrant pain, spontaneous rupture, infarction or thrombocytopenia.3,4 Currently, there are 21 case reports of spontaneous splenic rupture as a result of solid tumour metastases in the literature. We present the 22nd case, with a review of the previously reported cases and a discussion of the likely pathogenesis, clinical significance and prognostic implications of this rare but important condition.

CASE REPORT
A 54-year-old woman presented with an antral gastric carcinoma. The patient went on to have a distal gastrectomy with lymph node dissection, including hepatic artery and coeliac axis nodes. Histology revealed extension into the lesser omentum. Nine of 10 lesser curve and six of 11 oesophagogastric nodes were positive (total 15). The TNM stage was IIIB (T3N2M0). Nine months later, at routine follow up, recurrence was diagnosed in a left supraclavicular lymph node. Computed tomography (CT) scan identified extensive retroperitoneal, mesenteric, posterior mediastinal and porta hepatis lymphadenopathy. Palliative chemotherapy was commenced with good clinical and radiological response. No abnormality was noted in the spleen at that time or at three subsequent CT scans over the following 3 months. Seven months later, the patient developed a right-sided above-knee deep venous thrombosis (DVT) and was commenced on warfarin. Upon development of recurrent DVT in the same limb while therapeutically anticoagulated, warfarin was ceased and the patient commenced on subcutaneous low molecular weight heparin (1 mg/kg b.d.).
P. Smart MB BS; M. Cullinan MB BS; G Crosthwaite FRACS. Correspondence: Dr P. Smart, PO Box 4199, University of Melbourne, Vic. 3052, Australia. Email: maxand98@hotmail.com Accepted for publication 11 September 2001.

Seventeen months after the original diagnosis and surgery, the patient presented emergently with a 2-day history of gradual-onset left upper quadrant pain, which subsequently became severe. There was no history of recent trauma. While initially haemodynamically stable, the patient became acutely hypotensive and tachycardic in the emergency department. Initial full blood count revealed a haemoglobin of 81 g/L, normal white cell count and a platelet count of 42 109/L. Following resuscitation, haemoglobin had dropped to 57 g/L. Abdominal CT revealed an enlarged nonperfusing spleen, non-opacification of both splenic artery and vein and a large amount of free intraperitoneal fluid (Fig. 1). At laparotomy via a left upper quadrant incision, an ischaemic spleen was identified with spontaneous rupture at the hilum. Two litres of free intraperitoneal blood was drained. An emergency splenectomy was performed. No free intraperitoneal metastatic disease was noted. The spleen weighed 830 g and measured 190 110 80 mm. Macroscopic examination revealed multiple splenic ruptures extending from the hilum, in association with visible thrombus obstructing hilar vessels. A lake of blood 40 mm in maximum dimension was noted within the spleen. On microscopy, extensive diffuse infiltration of splenic red pulp by malignant cells of signet ring appearance with white pulp atrophy was seen. Malignant cells were identified in the splenic veins within the body of the spleen but not in extrasplenic vessels.

Fig. 1. Computed tomography (CT) of the abdomen showing non-perfusing spleen with free intraperitoneal fluid and nonopacification of splenic vessels.

154

SMART ET AL.

DISCUSSION
Spontaneous splenic rupture is a known complication of a number of pathological conditions. These include splenic cysts, various infectious diseases, including infectious mononucleosis, malaria and typhoid fever, or malignancy. Leukaemia and other haematological malignancies are responsible for the majority of cases,4 although spontaneous splenic rupture is a recognized but rare complication of solid tumour metastases. The tumours that most commonly metastasize to the spleen are melanoma (34%), breast carcinoma (12%), ovary (12%), lung (9%) and, less commonly, choriocarcinoma, prostate, oesophagus and endometrium.2,5 Splenic metastases are generally held to represent widely disseminated neoplastic disease, with the incidence of splenic involvement increasing up to 50% where four or more solid organs are involved simultaneously.3 However, cases of isolated metastases are increasingly being reported in parallel with advancements in diagnostic accuracy of imaging modalities and this may then herald an increase in the clinical prevalence of this condition.3 Primarily, metastases to the spleen from solid tumours manifest as nodules, with a diffuse pattern of infiltration usually associated with haematological malignancy.6 In a recent autopsy study of 72 patients with macroscopic splenic metastases, Lam and Tang7 reported the most common patterns of infiltration were as a solitary parenchymal nodule (42%) or multiple parenchymal nodules (41%) in roughly equal proportions. Occasionally, lesions involved only the splenic capsule (11%) and a diffuse pattern of infiltration is rarely seen (7%).7 Metastatic nodules found within splenic parenchyma are likely to represent haematogenous spread, whereas those found on the capsular surface are thought secondary to peritoneal seeding.8,9 There are 21 cases of spontaneous splenic rupture from metastases reported in the literature58,1025 (Table 1). Choriocarcinoma has been the most common primary tumour involved (five cases), followed by melanoma (four cases), lung (two large cell carcinomas, one adenocarcinoma, one squamous cell carcinoma), urinary bladder (two signet ring cell carcinomas, one undifferentiated), liver (one cholangiocarcinoma, one hepatocellular carcinoma), stomach (two cases) and rectal (one case). Mean patient age was 53 years (range 1984 years) and involved males more frequently (12 males, eight females, two unknown). There was no pattern of splenic infiltration identified as being more prone to rupture (five cases diffuse infiltration, five cases solitary nodule, five cases multiple nodules, six cases pattern unknown). In one case, no splenic tumour was identified, although the authors presumed microfoci of tumour responsible for rupture.15 No primary tumour was noted to have a predilection for a particular pattern of infiltration. Mortality was 100% in the five patients managed conservatively. One patient was treated angiographically and survived for the 4-month follow up. Of those treated surgically, five were long-term survivors (> 6 months). Six patients treated surgically died soon after splenectomy, although this was mainly related to other complications of the primary tumour. Survival period was unknown for seven patients. Therefore, the short-term prognosis of

Summary of reported cases of spontaneous splenic rupture from solid tumour metastases

Gastric Lung (adenocarcinoma) Choriocarcinoma Lung (large cell carcinoma) Gastric Rectal Choriocarcinoma Bladder (signet ring cell carcinoma) Lung (squamous cell carcinoma) Melanoma Choriocarcinoma Bladder (signet ring cell carcinoma) Cholangiocarcinoma Choriocarcinoma Choriocarcinoma Hepatocellular carcinoma Lung (large cell carcinoma) Melanoma Melanoma Gastric Bladder (undifferentiated carcinoma) Melanoma 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Present report 10 7 7 11 6 12 13 8 14 15 16 17 18 19 20 21 5 22 23 24 25 55/F 68/M 43/F 48/M N/A 41/M N/A/F 84/M 58/M 65/M 30/F 25/M 65/M 31/F 19/F 62/F 48/M 71/M N/S 77/F 67/M 51/M

Site of primary

Diffuse Solitary nodule N/S N/S N/A Diffuse N/A Diffuse Solitary nodule Multiple nodules No splenic tumour found Diffuse Multiple nodules Solitary nodule N/S Multiple nodules Solitary nodule Diffuse N/S Solitary nodule Multiple nodules Multiple nodules N/A, information not available; N/S, not stated by author/s.

Table 1.

Case no.

Reference

Age (years)/gender

Infiltration pattern

Surgical Surgical N/S N/S N/A Surgical Surgical Surgical Surgical Conservative Surgical Surgical Surgical Surgical Angiographic Conservative Surgical Conservative Conservative N/A Surgical Conservative

Management

25 days Long term (period N/S) N/S N/S N/A N/S (bone marrow infiltration) Long term (period N/A) N/S 77 days (cerebral metastasis) 2 days 14 days (cerebral metastasis) 6 months Long term (9 months follow up) Long term (3 years follow up) Long term (4 months follow up) 2 days 48 days 5 days N/S N/A 3 days 1 day

Survival period

Postoperative recovery was complicated by a large left middle cerebral artery territory stroke 2 days later and subsequent aspiration pneumonia. The patient died 23 days after presentation as a result of these complications.

SPONTANEOUS SPLENIC RUPTURE

155

these patients relates to the application of prompt surgical intervention. However, it is the prognosis of the primary tumour that determines life expectancy in the longer term. The pathogenesis of splenic rupture is multifactorial. Certainly splenomegaly, regardless of aetiology, predisposes to rupture, as demonstrated by this case series (average spleen weight 852 g). The mechanism of rupture also relates to the pattern of infiltration. With nodular infiltration, necrosis and cystic degeneration of the tumour mass leads to subcapsular haematomas, with subsequent rupture. Alternatively, the tumour mass may directly erode the splenic capsule. The mechanism of rupture in diffuse infiltration is less clear and probably relates to coagulopathy.5 Hypercoaguability may lead to thrombosis or embolism causing splenic infarction. Alternatively, hypocoagulable states, such as disseminated intravascular coagulation or thrombocytopenia, may result in catastrophic splenic haemorrhage in a fragile, pathologically enlarged spleen. There are a number of potential aetiological factors contributing to rupture in the case presented. Thrombocytopenia and an enlarged anatomically abnormal spleen are undoubtedly important. The absence of obvious infarction at pathological examination makes thromboembolism with subsequent infarction and rupture unlikely. However, a visible thrombus within splenic vessels suggests acute thrombosis may have led to a sudden increase in intrasplenic pressure, causing rupture in an already fragile spleen. Haemorrhage was then compounded by therapeutic anticoagulation for previous DVT.

CONCLUSIONS
Mortality in splenic rupture associated with haemodynamic instability is high in the absence of surgical intervention and, indeed, in the limited case series reviewed, conservative management of this condition produced a uniformly fatal outcome. The possibility of splenic rupture should be considered in patients with known metastatic neoplasm who present with unexplained circulatory collapse in association with left upper quadrant pain, but requires a high index of clinical suspicion. Splenectomy in this clinical scenario may palliate these patients successfully in the short term, although the advent of this condition portends a poor long-term prognosis. This prognosis relates mainly to that of the primary tumour rather than the presence of splenic metastases.

REFERENCES
1. Cummings OW, Mazur MT. Breast carcinoma diffusely metastatic to the spleen: A report of two cases presenting as idiopathic thrombocytopaenic purpura. Am. J. Clin. Pathol. 1992; 97: 4849. 2. Induhara R, Vogt D, Levin HS, Church J. Isolated splenic metastases from colon cancer. South. Med. J. 1997; 90: 6336. 3. Piardi T, DAdda F, Giampaoli F, Pulcini G, Lancini GP, Pouche A. Solitary metachronous splenic metastases: An evaluation of surgical treatment. J. Exp. Clin. Cancer Res. 1999; 18: 5758.

4. Cotran RS, Kumar V, Robbins SL (eds). Splenic tumour. In: Robbins Pathologic Basis of Disease, 5th edn. Philadelphia: Saunders, 1994; 66772. 5. Karakousis CP, Elias EG. Spontaneous (pathologic) rupture of spleen in malignancies. Surgery 1974; 76: 6747. 6. Acuthan R, Haray PN, Joseph A. Splenic metastases from a rectal tumour: An unusual presentation. Ann. R. Coll. Surg. 1999; 81: 139. 7. Lam KY, Tang V. Metastatic tumours to the spleen. A 25-year clinicopathologic study. Arch. Pathol. Lab. Med. 2000; 124: 52630. 8. Gupta PB, Harvey L. Spontaneous rupture of the spleen secondary to metastatic carcinoma. Br. J. Surg. 1993; 80: 613. 9. Lee SS, Morgenstern L, Phillips EH, Hiatt JR, Margulies DR. Splenectomy for splenic metastases: A changing clinical spectrum. Am. Surg. 2000; 66: 83740. 10. Massarweh S, Dhingra H. Solitary splenic metastasis in lung cancer with spontaneous rupture. J. Clin. Oncol. 2001; 19: 15745. 11. Natarajan P, Varshney S. Spontaneous splenic rupture secondary to metastatic gastric carcinoma. Trop. Gastroenterol. 2000; 21: 28. 12. Hou HC, Chen CJ, Chang TC, Hsieh TT. Metastatic choriocarcinoma with spontaneous splenic rupture following term pregnancy: A case report. Chang Gung Med. J. 1996; 19: 16670. 13. Ruther U, Schmidt A, Rupp W et al. Spontaneous splenic rupture in a patient with multiloculated signet ring cell carcinoma of the urinary bladder. Eur. Urol. 1993; 23: 41718. 14. Buzbee TM, Legha SS. Spontaneous rupture of spleen in a patient with splenic metastases of melanoma. A case report. Tumori 1992; 78: 478. 15. Giannakopoulos G, Nair S, Snider C, Amenta PS. Implications for the pathogenesis of aneurysm formation: Metastatic choriocarcinoma with spontaneous splenic rupture: Case report and a review. Surg. Neurol. 1992; 38: 23640. 16. Vidmar D, Baxter DL, Devaney K. Extensive dermal metastases from primary signet ring carcinoma of the urinary bladder. Cutis 1992; 49: 3248. 17. Al-Obaidi SM. Spontaneous rupture of the spleen due to metastatic carcinoma. Br. J. Clin. Pract. 1989; 43: 3856. 18. Kristoffersson A, Emdin S, Jarhult J. Acute intestinal obstruction and splenic haemorrhage due to metastatic choriocarcinoma: A case report. Acta Chir. Scand. 1985; 151: 3814. 19. Vujic I, Lutz MH, Curry N, Weinstein VJ. Angiographic management of bleeding in gestational trophoblastic malignancy. Am. J. Obstet. Gynecol. 1984; 149: 902. 20. Horie Y, Suou T, Hirayama C, Nagasako R. Spontaneous rupture of the spleen secondary to metastatic hepatocellular carcinoma: A report of a case and review of the literature. Am. J. Gastroenterol. 1982; 77: 8824. 21. Rydell WB, Ellis R. Spontaneous rupture of the spleen from metastatic carcinoma. JAMA 1978; 240: 534. 22. Turnbull A, Shah J, Fortner J. Recurrent melanoma of an extremity treated by major amputation. Arch. Surg. 1973; 106: 4968. 23. Szecker K, Geczev J, Buualka R. Spontaneous splenic rupture caused by cancer. Orv. Hetil. 1969; 1: 7912. 24. Goldstein HS. Spontaneous rupture of the spleen secondary to metastatic carcinoma. South. Med. J. 1966; 59: 2617. 25. Das Gupta T, Brasfield R. Metastatic melanoma: A clinicopathologic study. Cancer 1964; 17: 132339.

CORRESPONDENCE

617

Smart PJ, Kirton M, Mann GB: Cardiac tamponade during central venous access insertion. Australian and New Zealand Journal of Surgery. 2005;75:617.

Dear Editor,
Cardiac tamponade during central venous vascular access port insertion

Central venous catheters with subcutaneous injection ports are widely used as a means for delivering chemotherapy in cancer patients. They are frequently placed using the Seldinger technique, with a stiff plastic dilator and peel-away sheath used to position the catheter at the junction of the superior vena cava and right atrium. Complications associated with their insertion include pneumothorax, air embolism, catheter embolism, cardiac arrhythmias, arterial or venous injury, malposition and inadequate anchoring.1 After successful insertion there is an on-going risk of infection, occlusion, and thrombosis.1 A rare yet commonly fatal complication is cardiac tamponade, which carries a mortality of 6578%.1,2 Fluoroscopy to confirm the position of guidewire prior to dilation and placement of the catheter, and again to confirm the final position of the catheter, is recommended. Routine post-procedure chest radiography is also advised to document placement and to exclude pneumothorax.3 We present a case of acute haemopericardium and tamponade diagnosed intraoperatively using transoesophageal echocardiography (TOE) following insertion of a right subclavian central venous infusion port under fluoroscopic guidance. A 40-year-old woman developed liver metastases following resection of a rectal adenocarcinoma. There was no significant pericardial effusion on staging computed tomography (CT). Long-term venous access was required for chemotherapy. A Vortex VTX vascular access port (Horizon Medical Products,

GA, USA) was inserted into the right subclavian vein under general anaesthesia. The Seldinger technique was used to place a guidewire that was seen to lie in the cardiac silhouette on fluoroscopy. A dilator and peel-away sheath was placed over the guidewire and inserted to its full length. On removal of the guidewire and dilator, approximately 3 mL of serous fluid was noted to flow from the barrel of the peel-away sheath. Perforation of the pleura or the pericardium was suspected. The catheter was inserted and peel-away sheath was removed to allow contrast radiography to diagnose the problem. This showed contrast in the transverse sinus of the pericardium. A TOE probe was placed, which demonstrated the catheter in the transverse sinus behind the aorta anterior to the right pulmonary artery. There was associated pericardial fluid. The assistance of a cardiac surgeon was sought. With the patient prepped for a sternotomy and instruments ready, approximately 80 mL of heavily blood-stained fluid was aspirated via the infusaport under TOE vision. The catheter was then removed, and over the next 3 min blood recollected in the pericardial space and caused a cardiac tamponade. The patient became hypotensive (blood pressure falling from 90 to 95 mmHg to 6570 mmHg systolic) and circulatory support was required with metaraminol and adrenaline. A midline sternotomy was performed and a 23-mm puncture in the superior recess of the pericardium was repaired after evacuation of the tamponade. The patient made an uneventful recovery. Over 100 cases of cardiac tamponade caused by central venous catheters have been reported since 1954.1 Tamponade may occur acutely, or weeks after placement, and carries high mortality,1,2 largely attributed to missed diagnosis.4 Symptoms may be nonspecific and include dyspnoea, chest pain and dizziness. Becks triad of hypotension, raised venous pressure and muffled heart sounds is classically present. Pulsus paradoxus may also be found. During insertion, routine chest radiography, aspiration of venous blood and pressure transduction are recommended techniques to confirm correct placement, but are not failsafe and may not detect malposition.2 The guidewire over-insertion technique, where the guidewire is advanced into the right heart to provoke dysrhythmias, has been suggested as a more reliable means of placement, although with attendant increased risks of clinically significant arrhythmias. In the case described, the guidewire was seen to lie within the cardiac silhouette on fluoroscopy. However, we suspect that the guidewire moved proximally prior to dilation. The relatively rigid tip of the dilator may then have acted like a spear, and perforated the medial wall of the superior vena cava and the pericardium, instead of curving inferiorly towards the right heart along the guidewire. The lack of free aspiration of venous blood would have raised suspicion of misplacement, but the presence of a small amount of serous fluid flowing from the dilator alerted us to this possibility immediately. Where a guidewire, dilator or catheter is suspected to have perforated the pericardium, it must not be removed. It can act as a plug and/or drain until repair or drainage can proceed.5 In the present case this allowed us time to perform contrast radiology, arrange a TOE examination and enlist the assistance of a cardiac surgeon. The diagnostic modality of choice in suspected tamponade is echocardiography.5 Transoesophageal echocardiography is preferable if available because it does not contaminate the surgical field. To our knowledge this is the first reported case of cardiac tamponade as a result of central venous catheter placement diagnosed using on-table TOE. This allowed not only confirmation of

618

CORRESPONDENCE

the suspected diagnosis, but also removal of the dilator under vision, with real-time confirmation of an expanding haemopericadium prior to proceeding to sternotomy and repair. Although most systems incorporate soft wires and catheters, the reported case highlights the care that must be taken with relatively stiff dilators that may perforate the vein wall, especially given the acute angle of the path of the right subclavian vein as it joins the internal jugular vein to form the superior vena cava. It is critical that the wire is advanced sufficiently so that it protrudes beyond the tip of the dilator. Inserting the dilator under fluoroscopy may provide extra reassurance. It is also recommended that the dilator is not inserted to its full length, rather only so that the tip is within the access vein. The dilator should then be held steady and the sheath advanced over the well-introduced guidewire. Although rare it is important that the clinician is cognizant of the possibility of cardiac tamponade after central venous catheter placement.3 Echocardiography is the diagnostic modality of choice, and transoesophageal echocardiography in the acute setting allows on-table confirmation of the diagnosis and repair without contamination of the surgical field.

REFERENCES
1. Shields LB, Hunsaker DM, Hunsaker JC III. Iatrogenic catheterrelated cardiac tamponade: a case report of fatal hydropericardium following subcutaneous implantation of a chemotherapeutic injection port. J. Forensic Sci. 2003; 48: 41418. 2. Losert H, Prokesch R, Grabenwoger M et al. Inadvertent transpericardial insertion of a central venous line with cardiac tamponade: failure of preventive practices. Intensive Care Med. 2000; 26: 114750. 3. Collier PE, Sterling HB, Graff DM, Doyle P. Cardiac tamponade from central venous catheters. Am. J. Surg. 1998; 176: 21214. 4. Edwards H, King TC. Cardiac tamponade from central venous catheters. Arch. Surg. 1982; 117: 9657. 5. Forauer AR, Dasiker NL, Gemmete JJ, Theoharis CT. Pericardial tamponade complicating central venous insertions. J. Vasc. Interv. Radiol. 2003; 14: 2559.

PHILIP J. SMART, MB BS,* MICHAEL KIRTON, MB BS, FRANZCA, AND BRUCE MANN, MB BS, PHD, FRACS* Departments of *Surgery and Anaesthetics and Pain Management, Royal Melbourne Hospital, Melbourne, Victoria

Copymade by A onbeball of Datecopied

Surgical Practice (2005) 9, 147-149 Teaching corner

Axillopectoral muscle: An important anomaly in axillary surgery

Philip James Smart, Ramin Shayan and G. Bruce Mann*
Department of Surgery, Royal Melbourne Hospital, The University of Melbourne, Parkville, Victoria, Australia,

The axillopectoral muscle (Langer's axillary arch) is a common, though rarely noted, anatomical anomaly causing a range of rare clinical syndromes. Knowledge of the anomaly is increasingly important in the era of minimal access surgery and sentinel lymph node biopsy (SLNB) where it might produce false negative biopsies. We describe a case and review the current literature regarding its morphology. The axillopectoral muscle is found in 6% of human subjects, typically as a musculotendinous band 7-15 cm in length running between latissimus dorsi and pectoralis major with considerable morphological variation. it has been attributed to 2/15 false negative SLNB in one series. The axillopectoral muscle is common, variable and increasingly important in the era of SLNB. Key words: anat omy , axillary dissection, variation.

Introduction
In 1864, Ca r l Rit t er v o n Edenberg v o n Langer described an anomalous axillary muscle present in 4% of cadaveric dissections.' Subsequent studies reported the incidence of a 'Langer's arch' between 3 and 1 2 % . ' Although the anomaly is rarely noted in clinical prac t ic e, to cause axillary vein obstruction, 91 an axillary mass' -1 7 and tthe e hyperabduction s y ndrome. ' We describe a m h case encountered at axillary dissection for breast carL a n g e r cinoma and review the currently reported literature ' s regardingc morphology. We emphasize the imporits a r tance of this anomaly in an era of minimally invasive h surgery, where a thorough knowledge of anatomy and h a anatomic variations is paramount. s b e Case report e n A n 54-year-old woman underwent wide excision and o sentinel node biopsy for breast carcinoma, Pre- and t e intra-operative lymphatic mapping was carried out d using peritumoral technetium antimony sulphide and subareolar patent blue dye. The sentinel node was located without difculty through a 2. 5 cm incision below the axillary hairline. No anatomical abnormalities were suspected during this procedure. One of three sentinel nodes contained a metastasis so t he

patient returned the following week for completion axillary dissection. The previous incision was extended posteriorly, and an abnormal muscular structure was encountered early in the subsequent axillary dissection, whic h led t o temporary disorientation until t he anomaly was recognized. The muscle was immediately deep to subcutaneous fat, arose from a broad base on the anterior surface of latissimus dorsi and tapered as it passed antero-superiorly to the posterior aspect of pectoralis major. Here it merged through a ne tendon with the bres of pectoralis major approximately 10 cm from its insertion into the humerus. The abnormal muscle crossed the axillary vessels and brachial plexus anteriorly, measuring 2 cm at its base, and was approximately 7 cm in length (Figs 1, 2). A nerve supply was not seen. The muscle was divided and axillary dissection proceeded without further anatomical variation noted. Histological examination of the axillary specimen showed no metastases in 20 lymph nodes

Discussion
The axillopectoral muscle (Langer's axillary arch) is an anatomical variant found in approximately 6% of human subjects (Table 1). It is usually seen as a slender musculo-tendinous band or slip, 7-15 cm in length and 0.5-2.5 cm in width. It often arises from the lateral edge of latissimus dorsi at the mid-point of the posterior fold of the axilla and merges with the under surface of the tendon of pectoralis major. It crosses the axillary vessels and nerves anteriorly, and can be supplied by

"Author to whom all correspondence should be addressed. Email: brucesnann@mh.org.au Received 6 November 2004; accepted 21 December 2004,

148

PJ Smart et al,

Fig. 1. Operativ e photograph of the right axilla. pm, pectoralis major; br, right breast; apm, axillopectoral muscle; Id, latissimus dorsi.

Fig. 2. Sc hematic representation of an axillopectoral muscle. pm, pectoralis major; br, right breast apm, axillopectoral muscle; Id, latissimus dorsi.

Table 1. Inc idenc e and morphology of reported cases of axillopectoral ano a le s Author/year Langer 1864 Eisler 1912 Princteau 1915 Saitta 1962 Sachatello 1977 Boontje 1979 Le Double 1987 Weinzweig 1988 Perre 1989 Serpell 1991 Takafuji 1991 Ref 1 3 6 1 10 9 6 5 13 7 4 Age/sex NA NA NA 10 mol/L 15F 37F NA NS 54F 6OF 57 mol/L 93F 74F NS NS/M 48 mol/L 78F 70 mol/L 7OF Morphology NA NA NA 8 cm band from LD to PM PM to humerus Taut band from LD to PM NA NS Bilateral, otherwise NS 7 x lc m fusiform from LD to PM CB to pectoralis 'quartus' PM to CP Bilateral NS 3.5 x 8.5 cm from LD to CP, Pm, biceps and CB 1 1 x lc m 0.5 x 12cm to CP 15 x 0.7 cm from LD to PM 14 x 0.5 cm from serratus anterior to humerus and PM Nerve supply MP NA NA NS NS NS NA NS NS ICBN MP MP MP (left) TD (right) NS TD TD NS ID 6th intercostal Incidence ('/o) 4 7.7 12 NS NS NS 6 3 NS 0.25 6.4 Study type Cadaveric NA NA Clinical Clinical Clinical NA NS Clinical Clinical Cadaveric

Le Bouedec 1993 Dharap 1994 Yuksel 1996 Miguel 2001

6 12 14 2

1 NS NS 6

Clinical Cadaveric Cadaveric Cadaveric

CB, coracobrachialis; CP, coracoid process; ICBN, intercostobrachial nerve; LD, latissimus dorsi; MP, medial pectoral nerve; NA, not available; NS, not stated; PM, pectoralis major; Pm, pectoralis minor; Ref, reference number; TD, thoracodorsal nerve.

either the medial pectoral or thoracodorsal nerves in roughly equal proportions (Table 1). This variation in observed nerve supply supports the theory that the term 'Langer's arch' represents a group of, rather than single consistent, mus c ular anomaly in t he axilla.

There is considerable variation in the morphology of this muscle; it has been described arising from sorratus anterior or pectoralis major, and inserting ont o coracoid process of t he scapula, pectoralis minor, coracobrachialis, biceps brachii or the humerus.

The axillopectoral muscle

The muscle is thought to be a developmental remnant of the close functional and anatomical relationship between teres major and latissimus dorsi, and is formed as a result of a failure of apoptosis of muscle p r i mo r d i a . While 1 2 - 1 4 the muscle is largely clinically quiescent, the Langer's arch has been reported as causing axillary vein obstruction, 9 duction syndrome (characterized by positional venous a x i l l a r y oedema of the arm with minimal neuengorgement of m a or arterial involvement).' rologic s s l The Langer's arch is of greater signicance to the surgeon than as a pathological entity in its own right a n It d might inuence t he extent of axillary dissection, tespecially when not recognized early in the operation. h The e unexpected muscle can lead the surgeon to dissect above the axillary vein, thereby putting the cords h y p of r e the brachial plexus at risk and possibly increasing a the risk of lymphoeclema. It might also hide the lateral b group of axillary nodes, leading to an incomplete dissection with suboptimal staging and regional control of disease.' A recent series of localized axillary recurrence following SLNB attributed two of 15 recurrences to undissected axillopectoral muscles harbouring positive nodes . ' This recurrence was associated with signicant morbidity and is an independent predictor of metastatic disease. Giv en t hat t he mus c le might restrict entry into the axilla during axillary dissection as it is tautened by abduction and elevation of the arm, routine draping by wrapping the arm, thus allowing f ree mobilit y a n d relaxation o f t he mus c le i f encountered, is therefore advocated. When recognized, division of the muscle is recommended. I n theory, the presence of an axillary arch could precipitate axillary v ein entrapment o r ly mphoedema f ollowing latissimus dors i a p breas t reconstruction.' In the era of SLNB, small incisions limit the anatomical landmarks available to the surgeon. In the case described, the presence of a Langer's arch was not suspected and did not inuence the initial operation. In certain cases the sentinel node might lie deep to the Langer's arch causing difculty for the surgeon unaware of this variation,

Concl usi on The term Langer's arch represents a group of morphologically variable muscular anomalies present in 6% of the population. Although rarely directly pathological, recognition of its presence during axillary dissection i s vital. Restricted ac c es s i n S L NB means recognition of a Langer's arch is more challenging, but of equal importance so that the technique is carried out with optimum efcacy. References
1 Saitta OF, Baum V. Langer's axillary arch, an unusual cause of axillary mass. JAMA 1962; 180: 122, 2 Miguel M, Llusa M. Ortiz JC, Porta N, Lorente M, Gotzens V. the axillopectoral muscle (of Langer): report of three cases. Surg, Radio!. Anat. 2001; 23: 341-3. 3 Haagensen CD, Diseases of the Breast, 3rd edn. Philadelphia: W.B. Saunders Co., 1986. 4 Takafuji T, lgarashi J, Kanbayashi I etal. The muscular arch of the axilla and its nerve supply in Japanese adults, Kalbogashu Zasshi, 1991; 66: 511-23, 5 Weinzweig N, Browne EZ. Infraclavicular median nerve compression caused by a lipoma. Orthopedics 1988; 11: 10778, 6 Le Bouedec G, Dauplat J, Guillot M, Vanneuville G. Le muscle pectoro-axillaire. J. ChM 1993, 2: 66-9, 7 Serpell JW, ,Baum M. Signicance of 'Langer's axillary arch' in axillary dissection. ANZ J. Surg. 1991; 61: 310-12, 8 Sisley JF, The axillopectoral muscle, Surg, Gynecol. Obstet. 1987; 165: 73, 9 Boontje AH. Axillary vein entrapment. Br. J. Surg. 1979; 66: 331-2. 10 Sachatello CR. The axillopectoral muscle (Langer's axillary arch): a cause of axillary vein obstruction. Surgery 1977; 81: 610-12. 11 Hewitt RL, Acute axillary-vein obstruction by the pectoralisminor muscle. N. Engl. J. Med, 1968; 279: 595-6. 12 Dharap A. An unusually medial axillary arch muscle. J. Anat. 1994; 184: 639-41. 13 Perre Cl, Zoetmulder FAN, A bilateral axillopectoral muscle. Neth. J. Surg. 1989; 41: 49. 14 Yuksel M, Yuksel E, Surucu S. An axillary arch. Olin. Anat. 1996; 9: 252-4. 15 Wright FC, Walker J, Law CHI_ McCready DR, Outcomes after localised axillary node recurrence in breast cancer. Ann. Surg. Oncol. 2003; 10: 1054-8.

Smart PJ, Mann G B: Meta-analysis of techniques for closure of midline abdominal incisions (letter). British Journal of Surgery. 2003;90:370.
370 Correspondence

Pathogenesis and clinical management of hereditary non-polyposis colorectal cancer (Br J Surg 2002; 89: 13571369) Sir I was surprised that the authors did not take the opportunity to remind the readership of the key roles of hereditary bowel cancer registers and of the delivery of care in a multidisciplinary team setting1,2 . A prerequisite for dealing with these families is an accurate family pedigree. Few surgeons have the time to ascertain this and should therefore work closely with trained counsellors. The latter should also be available to deal with the many issues around genetic testing, as are well described in this paper. Not all surgeons can keep up to date with this rapidly changing eld. Thus while the educational intent of this paper is laudable, it lacks a suitable conclusion. Patients at risk of hereditary bowel cancer should be offered referral to clinicians with a special interest in this condition and who work in a multidisciplinary team. Finally, the value of registers (as part of the team) should receive acknowledgement in the areas of coordination of care, including surveillance of disparate organs, and gathering of data for education and research. A. D. Spigelman Faculty of Health, University of Newcastle, Callaghan, Newcastle, New South Wales 2310, Australia
DOI: 10.1002/bjs.4158

Meta-analysis of techniques for closure of midline abdominal incisions (Br J Surg 2002; 89: 13501356) Sir This attempt to provide clear evidence, especially regarding specic suture types and techniques, is commendable, although some further data are needed before accepting the authors recommendations. While simplication of effect estimates through meta-analysis is appealing, an assessment of the methodological quality of the studies analysed should have been included, in particular information on methods of blinding, allocation concealment, losses to follow-up, and whether analysis was based on intention to treat. Is there justication for the review in the light of two recent analyses? Subgroup analyses are prone to bias1 . Differences in length of follow-up can be controlled by meta-regression. The exclusion of studies of fewer than 100 patients seems counterintuitive; the strength of meta-analysis is the inclusion of such studies to give a reliable pooled effect estimate. Reporting of the search strategy so as to be reproducible would be preferable, with explicit reporting of methodology, and ideally some assessment of external validity. With meta-analyses proliferating, read by an increasingly non-medical readership, it is crucial that such analyses are conducted with rigorous and reproducible methodology. The value of narrative reviews should be to identify areas of uncertainty and therefore in need of further research. The perils of mistaken conclusions based on non-systematic reviews may be high, but those of reviews mistakenly labelled systematic must be higher. P. Smart and G. B. Mann Royal Melbourne Hospital, Grattan St Parkville, Melbourne 3052, Australia
DOI: 10.1002/bjs.4159

1 National Health and Medical Research Council. Guidelines for the Prevention, Early Detection and Management of Colorectal Cancer (CRC). NHMRC, 1999. http://www.health.gov.au:80/ nhmrc/publications/synopses/cp62syn.htm 2 Boardman MA. Heritable bowel cancer syndromes: recognition and preventive management. Gastroenterol Clin North Am 2002; 31: 11071131.

1 Smith GD, Egger M, Phillips AN. Meta-analysis. Beyond the grand mean? BMJ 1997; 315: 16101614.

Copyright 2003 British Journal of Surgery Society Ltd Published by John Wiley & Sons Ltd

www.bjs.co.uk

British Journal of Surgery 2003; 90: 367370

DOI:10.1111/j.1477-2574.2010.00265.x

HPB

LETTER TO THE EDITOR

Re: Alanine transaminase rather than abdominal ultrasound alone is an important investigation to justify cholecystectomy in patients presenting with acute pancreatitis

I write with reference to the informative study by Anderson and colleagues on the use of alanine transaminase (ALT) as an indicator of gallstone pancreatitis, published in a recent edition of this journal.1 Alkaline phosphatase (ALP) and g-glutamyl transferase (GGT) have been shown to be sensitive indicators for the presence of common bile duct stones and cholangitis;2,3 however, ALT is a more sensitive, but less specic, parameter. We read with interest Anderson et al.s description of the role of ALT in the treatment of acute pancreatitis1 and would like to resolve our subsequent queries and offer several comments. What was the reference standard for the diagnosis of gallstones? Did any patients with negative initial ultrasound (US) undergo a repeat interval US? Once the episode of acute pancreatitis has resolved, repeat US should become as sensitive for detecting gallstones as it is in subjects without pancreatitis (9298%).4,5 We would disagree with the statement that a negative US should be interpreted as a reason not to proceed to cholecystectomy. In our practice, patients with a negative initial US undergo subsequent US either prior to discharge or at the rst outpatient review. This eliminates the difculty involved in interpreting the US scan when ileus is present and increases sensitivity. When two US scans are negative, the patient does not drink signicant amounts of alcohol and rare causes are excluded, we would discuss cholecystectomy and cholangiography with any patient t for surgery. Investigations with low risk for morbidity that are designed to identify the main risk factor are imperfect, whereas those with a higher sensitivity carry costs and risks for morbidity that are not much lower than those of cholecystectomy and cholangiography. According to Anderson et al.,1 patients with alcoholic pancreatitis were excluded on the basis of history of alcohol consumption (>40 units of alcohol per week) and no US evidence of gallstones. This is contradictory, as the authors state that the sensitivity of US to detect gallstones in the setting of acute pancreatitis falls. Furthermore, patients with alcoholic pancreatitis may have incidental gallstones, whereas patients with gallstone pancreatitis may drink moderate amounts of alcohol. We perform US in patients with pancreatitis not only to nd gallstones, but to image the pancreatic head, liver texture and,

most importantly, the biliary tree. For patients with common bile duct stones, endoscopic retrograde cholangiopancreatography (ERCP) may represent the initial step in a management strategy. Other centres perform cholecystectomy with transcystic exploration of the bile duct (laparoscopic or open).6,7 In both instances, preoperative US is an integral step. Furthermore, some patients with pancreatitis, who are not suitable for cholecystectomy, may benet from ERCP or sphincterotomy, neither of which is performed without prior US.

Yahya Al-Habbal & Philip Smart

Department of Surgery Austin Health Heidelberg, Vic, Australia E-mail: yahya@alhabbal.info References 1. Anderson K, Brown LA, Daniel P, Connor SJ. (2010) Alanine transaminase rather than abdominal ultrasound alone is an important investigation to justify cholecystectomy in patients presenting with acute pancreatitis. HPB (Oxford) 12:342347. 2. Notash AY, Salimi J, Golfam F, Habibi G, Alizadeh K. (2008) Preoperative clinical and paraclinical predictors of choledocholithiasis. Hepatobiliary Pancreat Dis Int 7:304307. 3. Yang MH, Chen TH, Wang SE, Tsai YF, Su CH, Wu CW et al. (2008) Biochemical predictors for absence of common bile duct stones in patients undergoing laparoscopic cholecystectomy. Surg Endosc 22:16201624. 4. Cooperberg PL, Burhenne HJ. (1980) Real-time ultrasonography. Diagnostic technique of choice in calculous gallbladder disease. N Engl J Med 302:12771279. 5. Chak A, Hawes RH, Cooper GS, Hoffman B, Catalano MF, Wong RC et al. (1999) Prospective assessment of the utility of EUS in the evaluation of gallstone pancreatitis. Gastrointest Endosc 49:599604. 6. Fanning NF, Horgan PG, Keane FB. (1997) Evolving management of common bile duct stones in the laparoscopic era. J R Coll Surg Edinb 42:389394. 7. Williams EJ, Green J, Beckingham I, Parks R, Martin D, Lombard M. (2008) Guidelines on the management of common bile duct stones (CBDS). Gut 57:10041021.

HPB 2011

2011 International Hepato-Pancreato-Biliary Association

JSCR
Journal of Surgical Case Reports http://jscr.co.uk

Adult colonic intussusception: Surgery still the best option

Authors: M Dungerwalla, S Loh & P Smart Location: University of Melbourne, Melbourne, Australia Citation: Dungerwalla M, Loh S, Smart P. Adult colonic intussusception: Surgery still the best option. JSCR. 2012 6:3

ABSTRACT

Intussusception is the telescoping of proximal bowel wall into the lumen of a distal segment. Whilst it is common in children, intussusception in adults is rare, and predominantly occurs secondary to an underlying malignant neoplasm. Abdominal and pelvic computed tomography (CT) is preferred for detection of lead points and lesion localisation. We present the case of a 79-year-old female with a four-day history of colicky abdominal pain followed by obstipation and distension. CT demonstrated a rounded heterogeneous density protruding into the upper rectal lumen, and also left-sided colonic obstruction. Emergency laparotomy revealed a mid-sigmoid colonic mass intussuscepting into the rectum. Histopathology confirmed a T3N1 moderately differentiated colonic adenocarcinoma. Given the high likelihood of underlying malignancy, surgical reduction of the intussusceptum may be complicated by perforation and tumour spillage. En bloc resection using oncologic surgical principles remains the first line treatment.

INTRODUCTION

Intussusception is the telescoping of a proximal segment of bowel wall into the lumen of an adjacent, usually distal segment. It is commonly encountered in the small bowel of children, where the aetiology is largely benign and non-operative treatment usually successful (1). In contrast, intussusception in adults is rare (2). Colonic involvement represents up to 50% of cases, and in 70% the underlying cause is a malignant neoplasm (3). A wide range of other causes in the colon have been described including adenomatous polyps, inflammatory bowel disease, mycobacterial infection, and surgical anastomoses (4). Presentation is variable in adults, and commonly subacute. Presumably transient asymptomatic radiologic incidentalomas are more frequent, though these are usually seen in the small bowel (5). Those with symptoms report crampy abdominal pain, nausea, vomiting,

page 1 / 4

JSCR
Journal of Surgical Case Reports http://jscr.co.uk

and occasionally per rectal bleeding. Sigmoidorectal intussusceptions may be associated with diarrhoea and mucus. There may be symptoms of bowel obstruction, especially where the underlying cause is malignant (3). A palpable abdominal mass is infrequent (6). Accurate clinical diagnosis is difficult. Abdominal and pelvic computed tomography (CT) is the preferred test of choice with the advantage of detection of lead points and accurate localisation of the lesion (3,4). The findings are often pathognomonic and include a sausage shaped soft tissue mass with an eccentric fat density ring contained within, representing mesenteric fat. Mesenteric vasculature is often visible leading into the lesion (7). However increasing bowel wall oedema and progressive vascular compromise lead to a sequence of CT findings dependent on the time course of the intussusception. CT delineation of the underlying aetiology is unreliable, as a neoplastic lead point cannot be distinguished from the mass of an intussusception itself (3).

CASE REPORT

We present the following endoscopic and radiological images to highlight the clinical findings and reiterate the differences in the management of adult and paediatric cases. A 79-year-old female presented to the emergency department with a four-day history of intermittent, colicky abdominal pain. The patient reported a history of initial non-bloody diarrhoea with mucus, followed by obstipation and distension. Physical examination was consistent with a large bowel obstruction, revealing an uncomfortable patient with a distended abdomen and associated significant left iliac fossa tenderness. The rectum was empty on per rectal examination.

Subsequent CT scan revealed a rounded heterogeneous density protruding into the lumen of the upper rectum containing mesenteric fat and blood vessels (Figs. 1a-c). The appearance was suggestive of sigmoidorectal intussusception. There was significant associated proximal large bowel dilatation consistent with left-sided colonic obstruction. The patient proceeded to emergency surgery. On the operating table, flexible sigmoidoscopy confirmed the diagnosis of intussusception due to a sigmoidal tumour (Fig. 1d). At laparotomy, a mass in the mid sigmoid colon was found with intussusception of the tumour into the rectum.

page 2 / 4

JSCR
Journal of Surgical Case Reports http://jscr.co.uk

A Hartmanns procedure was performed. Post-operative recovery was complicated by rapid atrial fibrillation, but was otherwise uneventful. The patient was discharged home nine days after surgery. Histopathology of the surgical specimen showed a T3N1 moderately differentiated colonic adenocarcinoma.

DISCUSSION
This case highlights the key features of management of adult intussusception, and also demonstrates the differences to an approach towards paediatric intussusception. Due to the high rate of malignancy in adults with intussusception, pneumatic or hydrostatic reduction is ill advised given the risk of perforation, which is associated with higher peri-operative mortality and worse prognosis (8,9). A small number of cases of successful laparoscopic or endoscopic reduction in highly selected patient groups, with known underlying non-malignant aetiology have been reported (10). For the vast majority of presentations however, a malignant cause cannot be excluded. At surgery, reduction of the intussuscepted bowel segment should not be attempted due to the risk of tumour spillage. En bloc resection of the lesion using standard oncologic surgical principles remains the recommended first line treatment for adult colonic intussusception.

REFERENCES
1. Daneman A, Navarro O. Intussusception. Part 2: An update on the evolution of management. Pediatr. Radiol. 2004;34:97-108 2. Azar T, Taraneh MD, Berger DL. Adult intussusception. Ann. Surg. 1997;226:134-138 3. Huang BY, Warshauer DM. Adult intussusception: diagnosis and clinical relevance. Radiol. Clin. North Am. 2003;41:1137-1151 4. Floemer F, Bissig H, Oertli D, Bongartz G, Hamel CT. Multislice CT in adult colocolic intussusception: case report and review of the literature. Emerg Radiol. 2008;15:361366 5. Soni S, Moss P, Jaiganesh T. Idiopathic adult intussusception. Int J Emerg Med. 2011;4:8-11 6. Warshauer DM, Lee JKT. Adult intussusception detected at CT or MR imaging: Clinical-imaging correlation. Radiology. 1999;212:853-860 7. Gayer G, Zissin R, Apter S, Papa M, Hertz M. Adult intussusception a CT diagnosis. Br J Radiol. 2002;75:185-190 8. Zielinski MD, Merchea A, Heller SF, You N. Emergency management of perforated colon cancer: How aggressive should we be? J. Gastrointest. Surg. 2011; 15: 2232-2238

page 3 / 4

JSCR
Journal of Surgical Case Reports http://jscr.co.uk

9. Petersen VC, Baxter KJ, Love SB, Shepherd NA. Identification of objective pathological prognostic determinants and models of prognosis in Dukes' B colon cancer. Gut. 2002;51:65-69 10. Ho MM, Park JJ, Prasad LM. Post colonoscopy colonic intussusception reduced via a laparoscopic approach. JSLS. 2010;14:596-599

page 4 / 4
Powered by TCPDF (www.tcpdf.org)

ORIGINAL ARTICLE
ANZJSurg.com

Quantifying the cost of laparoscopic inguinal hernia repair

Philip Smart and Lindsay Castles
Department of Surgery, Austin Health, Heidelberg, Victoria, Australia

Key words cost analysis, health resources, inguinal hernia, laparoscopic surgery. Correspondence Dr Philip Smart, Department of Surgery, Austin Health, 145 Studley Road, Heidelberg, Vic. 3084, Australia. Email: philip.smart@austin.org.au P. Smart MBBS, FRACS; L. Castles, MBBS, FRACS. Accepted for publication 6 December 2011. doi: 10.1111/j.1445-2197.2012.06189.x

Abstract
Background: Laparoscopic inguinal hernia repair has been widely adopted. While the clinical advantages and disadvantages are well-quantied, the economic cost is less well-studied. Methods: A retrospective audit and case review of a single-centre public hospital consecutive case series was performed using data captured from real-time stock usage tracking technology and review of electronic medical records. Results: Laparoscopic inguinal hernia repair requires 11-min (21%) extra theatre time and costs $1268 (370%) more for single-use disposable equipment and prostheses. Conclusions: Laparoscopic inguinal hernia repair does not require signicantly more theatre time, but has a higher short-term in-hospital cost. Repair technique should be tailored according to patient factors. other parameters such as theatre usage times (hTrak, hTrak Pty. Ltd., Prahran, Australia). From November 2009, electronic medical records were adopted. These datasets were therefore reviewed and cross-checked for accuracy and completeness in order to perform this audit. Patients were identied by searching the hTrak database by procedure type. hTrak data are entered using a handheld device at the time of anaesthetic induction by theatre nursing staff. The data are then uploaded to a central, searchable SQL database accessible via web interface. The database contains a customizable dropdown menu for searching procedure types: the terms Inguinal hernia repair, Inguinal hernia, laparoscopic, unilateral/bilateral, Laparoscopic bilateral hernia repair, Laparoscopic hernia repair were available and selected. The platform is highly customizable and allows for inclusion of disease and procedure codes; however, these data were not included in the Surgery Centre implementation. Additional clinical data were obtained by reviewing the electronic medical record for scanned copies of handwritten anaesthetic charts, operation notes and outpatient notes as well as dictated outpatient letters. Patients with primary, recurrent, unilateral and bilateral herniae were all included. Patients with femoral herniae, or who had additional surgery during the same anaesthetic, were excluded.

Introduction
Over 26 000 inguinal hernia repairs are performed annually in Australia. Of these, 45% are performed laparoscopically, and this proportion has been steadily increasing.1 The risks and benets of laparoscopic inguinal hernia repair are relatively well-studied. A Cochrane meta-analysis demonstrated an equivalent recurrence rate but earlier return-to-work (6 days), albeit at the cost of higher rates of visceral and vascular injury.2 Chronic pain rates are lower at 1 year,2,3 but similar after 4 years.4 The economic dividend of the early return-to-work has been less wellstudied as has the higher cost of single-use disposable equipment. This study aimed to review the actual equipment usage and in-hospital cost of laparoscopic inguinal hernia repair versus open hernia repair using data from real-time tracking of stock usage, as well as quantifying differences in theatre utilization times.

Methods
All patients undergoing inguinal hernia repair at The Surgery Centre, Austin Health Repatriation Campus, from November 2009 to September 2011 were identied retrospectively. The Surgery Centre is a dedicated four-theatre facility designed for efcient and streamlined treatment of short-stay elective surgery patients. It was opened in July 2008, and is separate from the main Austin Health campus and Emergency Department. Various novel approaches were utilized in the facility design aimed at maximizing elective surgery efciency, including the adoption of point-of-use handheld mobile scanning data capture to manage stock usage and
2012 The Authors ANZ Journal of Surgery 2012 Royal Australasian College of Surgeons

Description of procedures
In the study population, surgery was generally performed by consultant surgeons, with the assistance of junior registrars. A small minority of cases were performed by senior registrars or fellows

ANZ J Surg (2012)

Smart and Castles

alone. Procedural technique was dictated by surgeon and/or patient preference during the standard operative consent process. In general, laparoscopic repair was not offered to patients with irreducible herniae or those with giant inguinoscrotal herniae. Prior appendicectomy or lower midline laparotomy was not considered a contraindication to laparoscopic repair. A small number of patients were only offered open repair due to co-morbidities precluding general anaesthesia. Mesh and mesh xation technique were dictated by surgeon preference. New prostheses and novel disposable equipment require approval by a committee prior to use, thus delaying the available mesh and xation device choices to some extent as compared to a wider population. For example, absorbable tacking devices were approved for use midway through the study period. Overnight beds were routinely booked as part of the pre-admission process, resulting in an almost universal overnight stay for both laparoscopic and open cases. While many patients may have been suitable for day case surgery, this was not yet routinely adopted during the study period because of the availability of extended evening recovery and re-admission processes.

drapes, mesh, mesh xation devices and theatre packs. Disposable ports were used routinely. Other than laparoscopic scissors that were used only occasionally, all other laparoscopic instruments were re-usable.

Statistical analysis
For normally distributed, continuous data, Students t-test was used to detect statistically signicant differences between groups. Categorical data were analysed using Fishers exact test. All P-values are two-tailed. A P-value <0.05 was considered statistically signicant.

Results
A total of 337 consecutive cases were identied. Six patients were excluded where the hernia was found to be femoral. A further 10 cases were excluded where additional surgery such as umbilical hernia repair was performed at the same time. Of the remaining 321 cases, the breakdown of hernia repair type was as follows: Totally extraperitoneal (TEP) Lichtenstein RutkowRobbins mesh plug Transabdominal preperitoneal 202 59 55 5 (64%) (18%) (17%) (1.5%)

Data collection and calculation of costs

Previous studies have used estimates for calculation of procedural costs.3,5 For this study, it was surmised that the main differences in in-hospital costs are due to differences in theatre utilization (i.e. laparoscopic cases take longer) and in the use of single-use disposable equipment. In our institution, general anaesthesia is used almost routinely, and therefore anaesthetic and preoperative work-up costs are equivalent. Patients also routinely stay overnight, regardless of surgery type. This allows comparison between two relatively hard endpoints: surgery time and actual disposables usage to estimate hospital costs. Theatre and surgery start and nish times are entered into the hTrak mobile handset by theatre nursing staff in real time. Disposables are barcode-scanned by the scout nurse as they are opened and used during a case. Disposables costs included a comprehensive list of single-use and consumable items ranging from dressings to sutures, disposable
Table 1 Baseline characteristics of open and laparoscopic repair groups Open Patients (n) Mean age (years) (range; SD) Age > 65 (n) (%) Gender (M/F) Weight (kg) (range; SD) ASA 1 ASA 2 ASA 3 ASA 4 Primary hernia Recurrent Right Left Bilateral General anaesthesia Local and sedation Overnight admission 114 61 (2090; 16.5) 54 (48%) 102/12 76 (49111; 13) 21 (18%) 63 (55%) 28 (25%) 2 (2%) 103 (90%) 11 (10%) 54 (47%) 50 (44%) 10 (9%) 105 (92%) 9 (8%) 111 (97%)

Although 22 individual surgeons were involved, four surgeons performed 84% of the cases. Most of the 18 remaining surgeons were either hospital surgeons from another campus covering the lists of regular surgeons on leave or were rotating fellows. Patients in the laparoscopic group were 9 years younger on average. Forty-eight per cent of patients in the open group were over 65 years old (i.e. of retirement age) compared to 18% in the laparoscopic group. They were also slightly heavier, but this 3-kg difference is not considered clinically signicant (Table 1). The open group had higher co-morbidity: more of these patients had an American Society of Anesthesiologists (ASA) score of 3, fewer had an ASA score of 1. A higher proportion of the laparoscopic group had a bilateral repair, but both groups included similar numbers of recurrent herniae.

Laparoscopic 207 52 (1783; 15.0) 37 (18%) 195/12 79 (43137; 11) 67 (32%) 124 (60%) 16 (8%) 0 (0%) 187 (90%) 20 (10%) 78 (38%) 59 (29%) 70 (34%) 207 (100%) 0 (0%) 194 (94%)

P-value 0.0001 0.0001 0.1276 0.025 0.0087 0.4781 0.0001 0.1254 1.0000 0.0001

ASA, American Society of Anesthesiologists; SD, standard deviation.

2012 The Authors ANZ Journal of Surgery 2012 Royal Australasian College of Surgeons

Cost of laparoscopic hernia repair

All laparoscopic cases were performed under general anaesthesia, as were most open operations. A few elderly patients had an open operation using local anaesthesia and sedation. Only 10% of cases in both groups were performed as day cases. Most repairs in the laparoscopic group were performed using contoured polypropylene mesh (Bard 3D max, Bard Medical, Covington, GA, USA; n = 116), followed by partially absorbable lightweight composite mesh (Ultrapro, Ethicon, Somerville, NJ, USA; n = 70). In 16 cases, the repair was performed using with standard polypropylene (Surgipro, Covidien, Dublin, Ireland), 3 using dual mesh (Proceed, Ethicon) and 2 with polyester (Parietex, Covidien). All mesh was xed, initially using 5-mm titanium tacks (Pro Tack, Covidien; n = 90) then once available mostly with absorbable synthetic polyester copolymer tacks (AbsorbaTack, Covidien; n = 101). Fibrin glue (Tisseel, Baxter, Deereld, IL, USA) was used for mesh xation in 16 cases. For open surgery, herniae were repaired with a preformed polypropylene plug (Bard PerFix Plug, Bard Medical; n = 48) where the RutkowRobbins technique was used. Either standard polypropylene (Surgipro, Covidien; n = 28) or partially absorbable lightweight composite mesh (Ultrapro, Ethicon; n = 38) was used for the Lichtenstein technique. Most mesh inserted at open surgery was xed with sutures (n = 96); however, 5-mm titanium tacks (Stat Tack, Covidien; n = 18) were used in some cases. Open cases were completed in 52 min on average, 11 min quicker than laparoscopic cases, which took 63 min (Table 2). Given the higher number of bilateral laparoscopic cases, unilateral and bilateral subgroups were then analysed separately. Unilateral open cases were completed in an average of 50 min, only 8 min faster than the laparoscopic repairs. Bilateral herniae were repaired 5 min faster using an open approach; however, given the low numbers of open cases, this did not reach statistical signicance. Total theatre usage time reected the longer surgery times alone: laparoscopic cases took only 13 min longer than open cases in total. This is the expected result given the almost universal use of general anaesthesia and serves to validate reliability of the data. Average cost of single-use and consumable equipment was $1611 for the laparoscopic group and $343 for the open group: a mean difference of $1268. Most of the cost difference was due to the use of laparoscopic tackers and brin xation, as well as the cost of disposable distension and structural balloons for establishing and maintaining the TEP space. This cost was stable over the 2-year

study period: mean cost for the rst 50 cases was $1586, for the last 50 mean cost was $1623 (P = 0.63).

Discussion
This study has some limitations. It is a retrospective single institution review of public hospital cases; therefore, the conclusions are not necessarily universally applicable. It is possible that some patients were missed in the database search as the operation descriptions are somewhat arbitrary and overlapping. For example, a case may be misclassied as an incisional hernia repair rather than inguinal hernia repair. This could be remedied by classifying the cases by procedure code in the database. The open and laparoscopic groups contained heterogeneity within them with regard to surgery type and style. We did not analyse these subgroups separately as we preferred to gain a broadly applicable overview of the open versus laparoscopic groups as a whole. In addition, the costs and time differences of the various styles of open repair were not felt to be signicant. The cost calculations in this study rely on the accuracy of the underlying database. In one instance, barcode scanning of a distension balloon recorded the use of a box of ve balloons rather than one single balloon, thereby inating the cost signicantly. This was corrected in the analysis, but there may be other instances that were missed. On the other hand, it is possible that some disposables were not scanned: in a busy theatre with many disposable items used, it is almost certain than some were missed, resulting in an underestimate of the true underlying cost. The cost estimates in this series do not take into account the additional capital expenditure required for laparoscopic equipment purchase and maintenance. Laparoscopic instrument trays are generally more time consuming to clean and sterilize. These factors would, of course, result in an underestimate of the true cost to the hospital of laparoscopic cases. On the other hand, laparoscopic cases may be cheaper in facilities where day case surgery is widely adopted, as these patients are likely to go home earlier. The time difference may have also been underestimated. Patients in the open group were older, heavier and sicker, all of which would lengthen operation times. In addition, the open group almost certainly included more patients with large inguinoscrotal herniae deemed not suitable for laparoscopic repair. These cases would also take longer and result in an underestimation of the extra surgical

Table 2 Cost of disposables and operating times Open Surgery time: all cases (min) (range; SD) Surgery time: bilateral herniae (min) (range; SD) Surgery time: unilateral herniae (min) (range; SD) Total theatre usage time (min) (range; SD) Disposables cost ($AUD) (range; SD)
CI, condence interval; SD, standard deviation.

Laparoscopic 63 (30150; 21) 77 (43150; 25) 58 (31124; 20) 96 (31211; 26) $1611 (7943188; 430)

Difference (95% CI) 11 (615) 5 (-199) 8 (313) 13 (619) $1268 (11861350)

P-value 0.0001 0.5 0.0009 0.0001 0.0001

52 (23129; 21) 72 (79132; 20) 50 (23129; 18) 108 (53192; 28) $343 (176964; 122)

2012 The Authors ANZ Journal of Surgery 2012 Royal Australasian College of Surgeons

Smart and Castles

time needed for laparoscopic cases. In a randomized trial, the time difference demonstrated in this study would likely be larger: indeed, in a meta-analysis of 7161 patients, the difference was 15 min in favour of open surgery.2 This study highlights the potential power of the implementation of information technology into clinical service areas. While this study examined two limited parameters (disposables cost and operating time), these can be easily broadened to capture unlimited endpoints. Dedicated data managers can be made obsolete as, with the correct implementation, automated data collection can now be integrated into normal clinical workows. Clinical benchmarks, identication of outlying performance, sentinel events and even trial data could potentially be gathered freely, as well as monitored and published in real time. These benets, however, can be accrued only with clinician involvement in the software design phase. A stock usage tracking system, with some customization, might be altered to aid identication of best surgical practice. Although not designed to do so, chart review of this case series also highlighted the higher rates of serious morbidity in the laparoscopic group. In this case series alone, it was noted that one patient required an emergency Hartmanns procedure for a sigmoid colon injury, one required emergency laparotomy to control bleeding from an iliopubic vein, two patients had a vas deferens division, one patient had an injury to the lateral cutaneous nerve of the thigh and two patients had a recurrent hernia within 30 days requiring re-operation. There were no major vascular or bladder injuries. These events are highly pertinent to cost analyses, as a serious complication would obliterate the cost benet of many other uncomplicated procedures. A more robust cost analysis would therefore require a comprehensive analysis of all complications and re-admission rates. In contrast among the open group, one patient who had had a previous paediatric herniotomy suffered a delayed thrombosis of the testicular veins, resulting in testicular ischaemia and atrophy. Another patient with a large inguinoscrotal hernia required orchidectomy at index hernia repair, and one patient in the open group required referral to a pain specialist for the management of chronic post-operative groin pain. No other major morbidity was noted in this older and frailer group and there were no early returns to theatre. These ndings are consistent with the reported literature. Open hernia repair is safer but at the cost of higher rates of chronic pain. In 2009, the European Hernia Society published comprehensive guidelines on the management of adult inguinal hernia based on an extensive literature review.4 It was noted that although the in-hospital costs were higher for laparoscopic repair, from a socioeconomic perspective, laparoscopic repair was probably the most cost-effective approach for patients who participate in the labour market, especially for bilateral hernias. Recommended repair technique should be in accordance with the published guidelines. Laparoscopic repair is recommended for bilateral hernias, recurrent hernias after previous anterior repair, for

patients in whom chronic pain is more likely or for working-age patients. It is safer to treat older or non-working patients with primary unilateral inguinal hernia by open or anterior technique. It is pleasing to note that this case series adhered to these recommendations in that patients in the laparoscopic group were younger and more often had bilateral herniae. In our series, the bilateral open repairs were generally performed in patients not t for general anaesthesia, and the recurrent herniae repaired via open technique either had a previous laparoscopic repair or were old and frail. Aside from early return-to-work, the other benet over open repair is a reduction in the prevalence of chronic pain. Risk factors for the development of chronic pain have been delineated and include recurrent hernia repair, the presence of preoperative pain, day case surgery, delayed onset of symptoms, high pain scores in the rst week after surgery, young age, preoperative anxiety level and occurrence of any post-operative complication.68 These risk factors should prompt consideration of laparoscopic over open repair.

Conclusion
Laparoscopic inguinal hernia repair costs $1268 more than open repair for disposable and single-use equipment, but only 11-min additional operating time on average. Laparoscopic repair is recommended in patients t for general anaesthesia with bilateral hernias, recurrent hernias after previous anterior repair, for patients in whom chronic pain is more likely or for working-age patients.

References
1. Medicare Australia Statistics. Medicare Item Reports. [Cited Oct 2011.] Available from URL: https://www.medicareaustralia.gov.au/statistics/ mbs_itemshtml. 2. McCormack K, Scott NW, Go PM, Ross S, Grant AM, EU Hernia Trialists Collaboration. Laparoscopic techniques versus open techniques for inguinal hernia repair. Cochrane Database Syst. Rev. 2003; 1: CD001785. 3. Liem MS, Halsema JA, van der Graaf Y, Schrijvers AJ, van Vroonhoven TJ. Cost-effectiveness of extraperitoneal laparoscopic inguinal hernia repair: a randomized comparison with conventional herniorrhaphy. Coala trial group. Ann. Surg. 1997; 226: 66875. 4. Simons MP, Aufenacker T, Bay-Nielsen M et al. European Hernia Society guidelines on the treatment of inguinal hernia in adult patients. Hernia 2009; 13: 343403. 5. Ibaez Rde M, Al Saied SA, Vallejo JA, Canales JM, Prieto CB, Sotos FE. Cost-effectiveness of primary abdominal wall hernia repair in a 364-bed provincial hospital of Spain. Hernia 2011; 15: 37785. 6. Poobalan AS, Bruce J, Smith WC, King PM, Krukowski ZH, Chambers WA. A review of chronic pain after inguinal herniorrhaphy. Clin. J. Pain 2003; 19: 4854. 7. Franneby U, Sandblom G, Nordin P, Nyren O, Gunnarsson U. Risk factors for long-term pain after hernia surgery. Ann. Surg. 2006; 244: 21219. 8. Macrae WA. Chronic post-surgical pain: 10 years on. Br. J. Anaesth. 2008; 101: 7786.

2012 The Authors ANZ Journal of Surgery 2012 Royal Australasian College of Surgeons

ORIGINAL ARTICLE
ANZJSurg.com

Thromboprophylaxis among Australasian colorectal surgeons

Philip Smart,* Kate Burbury, Senthil Lingaratnam, A. Craig Lynch, John Mackay and Alexander Heriot
*Department Department Department Department of of of of Cancer Surgery, Peter MacCallum Cancer Centre and Epworth Healthcare, Melbourne, Victoria, Australia Haematology, Peter MacCallum Cancer Centre, East Melbourne, Victoria, Australia Pharmacy, Peter MacCallum Cancer Centre, East Melbourne, Victoria, Australia and Cancer Surgery, Peter MacCallum Cancer Centre, East Melbourne, Victoria, Australia

Key words colorectal neoplasm, colorectal surgery, health care survey, thromboembolism. Correspondence Dr Philip Smart, Epworth Healthcare, 89 Bridge Road, Richmond, Vic. 3121, Australia. Email: phil.smart@petermac.org P. Smart MBBS, FRACS; K. Burbury FRACP, FRCPA; S. Lingaratnam BPharm, MPH; A. C. Lynch MMedSci, FRACS; J. Mackay FRACS, FRCS (Eng.); A. Heriot MD, FRACS. Accepted for publication 9 July 2012. doi: 10.1111/j.1445-2197.2012.06245.x

Abstract
Background: Thromboembolism is a common cause of morbidity and mortality in patients with colorectal cancer, but thromboprophylaxis (TP) is underutilized. Current guidelines do not make specic recommendations for colorectal cancer patients and provide minimal guidance for the ambulatory setting, although emerging evidence suggests TP may be warranted during chemoradiotherapy or in the extended postoperative phase. A survey of Australasian colorectal surgeons was therefore performed to assess current TP practice and attitudes. Methods: An online survey was sent to 204 surgeons who were members of the Colorectal Surgical Society of Australia and New Zealand. Results: One hundred twenty-eight surgeons (63%) completed the survey. Most surgeons consult available guidelines, and where recommendations are made, current practice is in line with them. Lack of data, lack of ownership, logistical issues and an absence of guideline recommendations currently prevent surgeons from instituting TP in the neoadjuvant treatment period. Fifty-four per cent of surgeons currently prescribe TP after hospital discharge; those that do not, cite logistical issues as the main constraint. Conclusion: More data on thromboembolism risk during various treatment phases are required and should be promulgated in tumour-specic guidelines. Logistical barriers to adopting TP in the ambulatory setting should be addressed.

Introduction
Each year, an estimated 30 000 Australian hospitalizations relate to thromboembolism (TE) and 5000 patients die as a result.1 Malignancy is an independent risk factor for the development of TE, with an incidence rate of up to 1 in 200 per year in patients with cancer,2 and this rate is also increasing.3,4 TE in cancer patients is associated with signicant clinical consequences including a 2.2-fold increase in mortality, development of post-thrombotic syndromes and pulmonary hypertension, bleeding complications related to anticoagulant therapy, increased health resource utilization and also contributes to treatment delays or omissions.57 Although the highest rates are seen in ovarian, brain and pancreatic cancer, when adjusted for disease prevalence, the commonest malignancies associated with thrombosis are those of the colon, breast and lung.2,8 In one cohort of 68 142 colorectal
2012 The Authors ANZ Journal of Surgery 2012 Royal Australasian College of Surgeons

cancer patients, 5% developed a TE in the rst 6 months following diagnosis.9 Thromboprophylaxis (TP) safely and cost-effectively reduces TE by up to 70%, but remains signicantly underutilized.10,11 It is not known to what extent the lack of data specic to individual tumour streams or the known underrecognition of both the prevalence and morbidity of TE by treating clinicians contributes to underprescribing TP.12,13 Modern management of colorectal cancer is multidisciplinary and includes neoadjuvant chemoradiotherapy (nCRT), surgical resection, adjuvant and palliative chemotherapy, and radiotherapy, all of which may confer an additional risk for the development of TE. These treatments are increasingly being delivered in the ambulatory care setting and this has led to interest in expanding the TP treatment period to encompass both the nCRT and extended post-operative phases (extended TP).14 A 2009 Cochrane systematic review

ANZ J Surg (2012)

Smart et al.

demonstrated that extended TP after abdominal or pelvic surgery was associated with reduction in TE at 30 days from 14.3% to 6.1% (P < 0.0005).15 A number of international and national guidelines outlining optimal TP practice have been produced,1619 including some that are cancer-specic.2024 However, most TP guidelines currently acknowledge a lack of data in the setting of nCRT, TP and extended post-operative phases and offer only general recommendations for consideration of TP in high-risk patients. TE risk and optimal TP practice during various treatment phases specic to colorectal cancer are therefore yet to be dened. Given the lack of specic recommendations, particularly in the ambulatory care setting, we conducted a survey of the attitudes and current practices of colorectal surgeons in Australia, with a particular focus on TP during the nCRT and extended post-operative phases.

not use them largely because they disagreed with the recommendations or felt they were not relevant. Of those who had access to formal guidelines, 60% reported that TP recommendations applied to inpatients only, 18% reported that the guideline contained separate recommendations for inpatients and outpatients, and 17% reported that the guideline did not make the distinction between inpatient and outpatient groups.

Mechanical prophylaxis
The majority (95%) of respondents used thromboembolic deterrent stockings. Ninety-four per cent used calf compressors and 90% early mobilization. One respondent used foot pumps.

P-TP
Ninety-eight per cent of respondents routinely used P-TP. The majority (66%) utilized enoxaparin, with the most common dose being 40 mg daily, commenced within 6 h post-op and continued until discharge from hospital. A smaller number of respondents used standard unfractionated heparin (30%), and commenced the agent on anaesthetic induction (Fig. 1).

Methods
Permission to survey members was granted from the Colorectal Surgical Society of Australia and New Zealand (CSSANZ) who identied 190 practising members and 23 trainees. An email containing a link to a 20-question online survey was sent to members in February 2012. The survey questions outlined in Table 1 focused on TP practices including the presence of hospital or unit guidelines, TP modalities and prescriptions, as well as the timing and duration of pharmacological TP (P-TP). The survey was reviewed by a multidisciplinary team of clinicians comprising colorectal surgeons, a haematologist, a medical oncologist and a pharmacist prior to being piloted within a single institution to ensure validity and reliability of responses. The electronic online survey incorporated logic that channelled follow-on questions depending on prior responses; therefore, not all questions were answered by each respondent. Duplicate responses were avoided by allowing one computer to complete one survey only and allocated three-digit codes for each member through the use of unique CSSANZ. The ordering of available answers was randomized among the users to decrease bias. Non-responders were sent a reminder email after 2 weeks. If no response was received 2 weeks after the reminder email, non-participation was assumed. All responses were anonymous and no demographic data was requested. Answers to questions were recorded as given and analysed online using SurveyMonkey (Palo Alto, CA, USA).

Prophylaxis during nCRT

Almost half (47%) of respondents considered patients receiving nCRT to be at higher risk of TE, with the remaining (40%) not aware as to whether patients were at risk and 13% did not consider patients at increased risk. However, the majority of respondents (81%) did not recommend P-TP during this period, while 4.5% recommended P-TP for all patients and 14% recommended for selected cases only. Of the respondents who recommended P-TP, 62% suggested treatment should continue for the duration of nCRT. However, the choice of agent was left up to the radiation (33%) or medical (24%) oncologist. Of those who took ownership of the P-TP, 19% would use enoxaparin, 5% unfractionated heparin, 5% aspirin, 5% an oral antithrombin agent and 5% switched between dalteparin and enoxaparin. Nineteen per cent considered the appropriate duration for P-TP during nCRT to be 14 weeks from the beginning of treatment. Fourteen per cent used P-TP continuously from the commencement of nCRT through to the post-operative phase while 5% used it for less than 1 week from the start of treatment. Those who did not recommend P-TP during nCRT were asked for reasons, with more than one selected response allowed. The majority (45%) were not aware of evidence supporting the use of P-TP during this phase of treatment, while 27% did not think the data was robust and/or felt it was the responsibility of the medical (31%) or radiation oncologist (33%). Twenty-ve per cent of the respondents thought the treatment was logistically difcult to arrange and 26% did not prescribe P-TP during nCRT as it was not recommended by their hospital guidelines.

Results
Response rates
Of the 204 CSSANZ members identied, 128 (63%) responded. Those who responded completed all questions in full.

Guidelines
Sixty-three per cent of respondents reported access to unit- or hospital-based guidelines for the prevention of TE in colorectal cancer patients. When guidelines were available, the majority (80%) were compliant and referred to them frequently in clinical practice, while 11% used the guidelines sometimes and a minority (8%) did

Extended P-TP
The majority of respondents (47%) recommend P-TP for selected patients after hospital discharge, 45% did not recommend it for any patient and 8% recommended P-TP for all patients with colorectal cancer after discharge. Of those who recommended P-TP, nearly all utilized lowmolecular-weight heparin (LMWH) (97% enoxaparin,
2012 The Authors ANZ Journal of Surgery 2012 Royal Australasian College of Surgeons

Thromboprophylaxis in colon cancer

Table 1 Survey questions Question Does your unit or hospital carry a formal protocol or pre-printed guidelines for prevention of DVT/PE in colorectal cancer patients? Do you use them? Yes/No/Do not know Potential responses

Which patients do the guidelines apply to? Which methods of mechanical prophylaxis do you routinely use? (You can choose more than 1 response) Do you routinely use pharmacological prophylaxis for SURGICAL INPATIENTS? Which agent do you use? Why not? What is your typical prescription? When do you START pharmacological prophylaxis? When do you routinely CEASE pharmacological prophylaxis? Do you think patients treated with neoadjuvant chemoradiotherapy are at increased risk of DVT/PE in the NEOADJUVANT CRT treatment period? Do you recommend pharmacological thromboprophylaxis for patients having NEOADJUVANT CHEMORADIOTHERAPY? When do you use it?

Always/Sometimes/Never No because I disagree with the recommendations/No because they are out of date No because they are general guidelines not relevant to my patients No because they are hard to access when I need them All patients as a generic group/Inpatients only/Both Inpatients and Outpatients/Do not know TED stockings/Calf compressors/Foot pumps/Early mobilization (<24 h)/None Yes/No Standard unfractionated heparin/Enoxaparin (Clexane)/Dalteparin (Fragmin) Other LMWH/Aspirin/Oral antithrombin agent (not warfarin)/Warfarin/Other (please specify) Not superior to mechanical methods/Not safe: too much bleeding Unnecessary: risk of DVT/PE low/Ineffective/Inconvenient to administer/monitor/Other (specify) Heparin 2500 units/Heparin 5000 units Enoxaparin 20 mg/Enoxaparin 30 mg/Enoxaparin 40 mg/Enoxaparin 60 mg Dalteparin 2500 units/Dalteparin 5000 units/Dalteparin 7500 units/Dalteparin 10 000 units Day prior to surgery/Morning of surgery/On induction/At completion of surgery/Within 6 h post-op/612 h. post-op/Within 24 h post-op/After 24 h pos-op When mobile/On discharge/<1 week post-op/46 weeks post-op/>6 weeks post-op Yes/No/Do not know

Yes, for selected patients/Yes, for all patients/No

Which agent do you use?

What is your typical prescription? Why not? (can select more than one option)

Do you recommend post-operative pharmacological thromboprophylaxis AFTER HOSPITAL DISCHARGE (i.e. extended prophylaxis)? Which agent do you use? What is your typical prescription? Why not? (can select more than one option)

For <1 week (from the beginning of treatment)/For 14 weeks (from the beginning of treatment) For the duration of neoadjuvant chemoradiotherapy Continuously from the commencement of neoadjuvant therapy through to post-op phase, and only interrupt in the peri-op period I leave it up to the Radiation Oncologist/I leave it up to the Medical Oncologist Standard unfractionated heparin/Enoxaparin (Clexane)/Dalteparin (Fragmin) Other LMWH/Aspirin/Low molecular weight heparin (LMWH) Oral antithrombin agent (not warfarin)/Warfarin/Other (please specify) Heparin 2500 units/Heparin 5000 units Dalteparin 2500 units/Dalteparin 5000 units/Dalteparin 7500 units/Dalteparin 10 000 units Enoxaparin 20 mg/Enoxaparin 30 mg/Enoxaparin 40 mg/Enoxaparin 60 mg I do not believe there is an increased risk/I am aware there may be increased risk, but I am not aware of any data/I do not think the data is robust It is the responsibility of the radiotherapist/It is the responsibility of the medical oncologist It is not required as DVT/PE risk is mainly in the immediate post-op period Patient compliance with subcutaneous injection is low It is logistically difcult to arrange/I am concerned about bleeding I am concerned about side effects e.g. HITS/It is too expensive for patients Theres no suitable oral agent/My hospital guidelines do not recommend it Yes, for all patients/Yes, for selected patients/No

Standard unfractionated heparin/Enoxaparin (Clexane)/Dalteparin (Fragmin) Other LMWH/Aspirin/Oral antithrombin agent (not warfarin)/Warfarin/Other (please specify) Heparin 2500 units/Heparin 5000 units Enoxaparin 20 mg/Enoxaparin 30 mg/Enoxaparin 40 mg/Enoxaparin 60 mg Dalteparin 2500 units/Dalteparin 5000 units/Dalteparin 7500 units/Dalteparin 10 000 units It is not required as DVT/PE risk is mainly in the immediate post-op period/I am not aware of any evidence in favour of it/Poor patient compliance with subcutaneous injection/It is logistically difcult to arrange/I am concerned about bleeding/I am concerned about side effects (e.g. HITS) It is too expensive for patients/I would if there was a suitable oral agent/My hospital guidelines do not recommend it

CRT, chemoradiotherapy; DVT, deep vein thrombosis; HITS, heparin-induced thrombocytopaenia; PE, pulmonary embolism; TED, thromboembolic deterrent stocking.

2012 The Authors ANZ Journal of Surgery 2012 Royal Australasian College of Surgeons

Smart et al.

Fig. 1. Commencement of thromboprophylaxis in surgical inpatients.

Fig. 2. Reasons for not recommending extended thromboprophylaxis. DVT, deep vein thrombosis; HITS, heparininduced thrombocytopaenia; PE, pulmonary embolism.

2% dalteparin) with a standard dosing schedule of enoxaparin 40 mg daily prescribed by the majority (96%) of those who used it. The most common reason given for not prescribing P-TP was that it was logistically difcult to arrange (50%), lack of awareness of any supporting evidence (31%), poor patient compliance (17%) or that hospital guidelines did not recommend it (40%) (Fig. 2).

Discussion
This survey of Australasian specialist colorectal surgeons with regard to TP in colorectal cancer patients has found that current practice is generally in line with recommended guidelines. However, it has also found that there is heterogeneity with regard to practice in the ambulatory setting. The clinical uncertainty regarding the utility of TP during the nCRT and extended postdischarge phases is reected in both the guideline recommendations and surveyed population. The TE risk and optimal TP practices need to be claried for both tumour groups and different treatment strategies. In particular, the highest risk periods should be carefully dened, and the riskbenet ratios of LMWH in the ambulatory care setting delineated. Many surgeons felt that logistical concerns prevented them from instituting TP during nCRT or in the extended post-operative TP phase. This highlights a need to integrate ambulatory, inpatient and primary physician care through enhanced communication channels and increased resourcing of outreach services and patient education programmes. Although formal cost-effectiveness analysis has yet to be conducted in this patient population, there is likely to be signi-

cant cost-effectiveness for adequately resourced P-TP programmes from both societal and hospital perspectives.25 Although both UK and Australasian colorectal surgeons are likely to use a guideline (80% versus 94%), Australasians are less likely to use LMWH compared with unfractionated heparin (68% versus 94%). UK surgeons start P-TP earlier (68% versus 18% on admission), but also cease it earlier (18% versus 4% prior to discharge). The survey had a response rate of 62%. A response rate over 60% has been accepted as the threshold above which survey responses can be extrapolated to the target population as a whole.26 This response rate compares well with similar surveys of TP practice in colorectal (53%) and orthopaedic surgery (55%).27,28 However, the response rate may have been higher had a mailed version been used for email non-responders. There are limitations in extrapolating the survey data to real world practice. An interested respondent who is likely to prescribe TP at higher rates would be overrepresented in our sample. Nonresponders may be less likely to adhere to current recommendations, thus distorting the data. Equally, although consultant surgeons and/or fellows may be committed to developing unit policy regarding TP, they may not be involved with day-to-day prescribing of P-TP and so their attitudes may not reect what occurs in practice. Cross-sectional surveys, such as the ENDORSE study that measured actual prescribing habits through a chart review of inpatients across 32 countries, demonstrated American College of Chest Physician guideline compliance rates of 72.3% for an undifferentiated group of patients undergoing major surgery for any cancer.11 This gure is signicantly lower than that suggested by our survey.
2012 The Authors ANZ Journal of Surgery 2012 Royal Australasian College of Surgeons

Thromboprophylaxis in colon cancer

Conclusion
TP practice by Australasian CSSANZ members is in line with current guidelines. There is a need to clarify TE risk in colorectal cancer during the nCRT treatment period as well as to clarify the evidence for extended TP post-discharge. Tumour-specic TE guidelines should be developed. Many surgeons view logistical barriers as an important factor in instituting extended TP.

Acknowledgements
We thank the members of the CSSANZ for their involvement, and Ms Liz Neilson for providing the list of active members of the association and disseminating the survey.

References
1. National Health and Medical Research Council. Stop the Clot: Integrating VTE Prevention Guideline Recommendations into Routine Hospital Care, 3rd edn. Melbourne: National Health and Medical Research Council, 2011. 2. Lee AY, Levine MN. Venous thromboembolism and cancer: risks and outcomes. Circulation 2003; 107: I1721. 3. Stein PD, Beemath A, Meyers FA, Skaf E, Sanchez J, Olson RE. Incidence of venous thromboembolism in patients hospitalized with cancer. Am. J. Med. 2006; 119: 608. 4. Khorana AA, Francis CW, Culakova E, Kuderer NM, Lyman GH. Frequency, risk factors, and trends for venous thromboembolism among hospitalized cancer patients. Cancer 2007; 110: 233946. 5. Lyman GH. Venous thromboembolism in the patient with cancer: focus on burden of disease and benets of thromboprophylaxis. Cancer 2011; 117: 133449. 6. Khorana AA. Cancer and thrombosis: implications of published guidelines for clinical practice. Ann. Oncol. 2009; 20: 161930. 7. Sorensen HT, Mellemkjaer L, Olsen JH, Baron JA. Prognosis of cancers associated with venous thromboembolism. N. Engl. J. Med. 2000; 343: 184650. 8. Levitan N, Dowlati A, Remick SC et al. Rates of initial and recurrent thromboembolic disease among patients with malignancy versus those without malignancy. Risk analysis using Medicare claims data. Medicine 1999; 78: 28591. 9. Alcalay A, Wun T, Khatri V et al. Venous thromboembolism in patients with colorectal cancer: incidence and effect on survival. J. Clin. Oncol. 2006; 24: 11128. 10. Mismetti P, Laporte S, Darmon JY, Buchmuller A, Decousus H. Metaanalysis of low molecular weight heparin in the prevention of venous thromboembolism in general surgery. Br. J. Surg. 2001; 88: 91330. 11. Kakkar AK, Cohen AT, Tapson VF et al. Venous thromboembolism risk and prophylaxis in the acute care hospital setting (ENDORSE survey): ndings in surgical patients. Ann. Surg. 2010; 251: 3308. 12. Ryu JH, Olson EJ, Pellikka PA. Clinical recognition of pulmonary embolism: problem of unrecognized and asymptomatic cases. Mayo Clin. Proc. 1998; 73: 8739.

13. Mandala M, Barni S, Floriani I et al. Incidence and clinical implications of venous thromboembolism in advanced colorectal cancer patients: the GISCAD-alternating schedule study ndings. Eur. J. Cancer 2009; 45: 6573. 14. Merkow RP, Bilimoria KY, McCarter MD et al. Post-discharge venous thromboembolism after cancer surgery: extending the case for extended prophylaxis. Ann. Surg. 2011; 254: 1317. 15. Rasmussen MS, Jorgensen LN, Wille-Jorgensen P. Prolonged thromboprophylaxis with low molecular weight heparin for abdominal or pelvic surgery. Cochrane Database Syst. Rev. 2009; (1): CD004318. 16. Stahl TJ, Gregorcyk SG, Hyman NH, Buie WD. Practice parameters for the prevention of venous thrombosis. Dis. Colon. Rectum. 2006; 49: 147783. 17. Lyman GH, Khorana AA, Falanga A et al. American Society of Clinical Oncology guideline: recommendations for venous thromboembolism prophylaxis and treatment in patients with cancer. J. Clin. Oncol. 2007; 25: 5490505. 18. Geerts WH, Bergqvist D, Pineo GF et al. Prevention of venous thromboembolism: American College of Chest Physicians EvidenceBased Clinical Practice Guidelines (8th Edition). Chest 2008; 133: 381S453S. 19. National Health and Medical Research Council. Clinical practice guideline for the prevention of venous thromboembolism (deep vein thrombosis and pulmonary embolism) in patients admitted to Australian hospitals. Melbourne, Australia: National Health and Medical Research Council, 2009. 20. Barni S, Falanga A, Mandala M, Piccioli A. Tromboembolismo venoso nei pazienti con tumori solidi. Edition. Italy: Associazione Italiana di Oncologia Medica, 2010. [Cited 13 Nov 2011.] Available from URL: http://www.aiom.it/Attivit%E0+Scientica/Linee+guida/Archivio+ 2010/Tromboembolismo+venoso+nei+pazienti+con+tumori+solidi/ 1,4770,0, 21. Streiff MB, Bockenstedt PL, Cataland SR et al. Venous thromboembolic disease. J. Natl Compr. Canc. Netw. 2011; 9: 71477. 22. Siragusa S, Armani U, Carpenedo M et al. Prevention of venous thromboembolism in patients with cancer: guidelines of the Italian Society for Haemostasis and Thrombosis (SISET). Thromb. Res. 2011; 129: 1716. 23. Mandala M, Falanga A, Roila F. Management of venous thromboembolism (VTE) in cancer patients: ESMO Clinical Practice Guidelines. Ann. Oncol. 2011; 22(Suppl 6): vi8592. 24. Khorana AA, Streiff MB, Farge D et al. Venous thromboembolism prophylaxis and treatment in cancer: a consensus statement of major guidelines panels and call to action. J. Clin. Oncol. 2009; 27: 491926. 25. Access Economics. The burden of venous thromboembolism in Australia. Report for The Australia and New Zealand Working Party on the Management and Prevention of Venous Thromboembolism, 2008. 26. Livingston EH, Wislar JS. Minimum response rates for survey research. Arch. Surg. 2012; 147: 110. 27. Srinivasaiah N, ArsalaniZedah R, Monson JR. Thrombo-prophylaxis (TP) in colo-rectal surgery: a National Questionnaire Survey (NQS) of the members of the ACPGBI. Col. Dis. 2012; 14: 3903. 28. Molnar RB, Jenkin DE, Millar MJ, Campbell D, Harris IA. The Australian arthroplasty thromboprophylaxis survey. J. Arthroplasty 2012; 27: 1739.

2012 The Authors ANZ Journal of Surgery 2012 Royal Australasian College of Surgeons