Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
9
Copyright © 1997 by The Johns Hopkins University School of Hygiene and Public Health Printed in U.S.A.
All rights reserved
Mike S. L. Liem,1 Yolanda van der Graaf,2 Reinder C. Zwart,1 Ingrid Geurts,1 and
Theo J. M. V. van Vroonhoven1 on behalf of the Coala Trial Group3
Potential risk factors for inguinal hernia in women were investigated and the relative importance of these
factors was quantified. In women, symptomatic but nonpalpable hernias often remain undiagnosed. However,
knowledge on this subject only concerns hernia and operation characteristics, which have been obtained by
review of case series. Virtually nothing is known about risk factors for inguinal hernia. The authors performed
a hospital-based case-control study of 89 female patients with an incident inguinal hernia and 176 age-
matched female controls. Activity since birth with two validated questionnaires was measured and smoking
habits, medical and operation history, Quetelet index (kg/m2), and history of pregnancies and deliveries were
recorded. Response for cases was 81 % and for controls 73%. Total physical activity was not associated with
inguinal hernia (univariate odds ratio (OR) = 0.8, 95% confidence interval (Cl) 0.6-1.1), but high present sports
family characteristics; hernia, inguinal; obesity; physical fitness; sports; women's health
Inguinal hernia in females, although not as common Indeed, this explains the lower incidence in females,
as in males, still results in an appreciable number of but it does not explain why some females do acquire
operations yearly. For instance, in the Netherlands in an inguinal hernia.
1994, there were approximately 3,500 operations for Inguinal hernias can either be congenital or ac-
inguinal hernia in females out of a total of 33,000 quired. In predominantly male populations, the risk
operations (1). Reports of case series of female ingui- factors that have been found to be associated with
nal hernia, and that have been mostly uncontrolled and inguinal hernia are muscle deficiency (previous appen-
retrospective, have emphasized the different, but fa- dectomy or other abdominal operations), physical
vorable, stronger inguinal anatomy in females (2-4). stress, intra-abdominal pressure (obstipation and pros-
tatism), smoking, aging, pelvic fractures and trauma,
connective tissue disease, and systemic illnesses (5-10).
Received for publication January 13, 1997, and accepted for In females, obesity, pregnancy, and operative pro-
publication June 26, 1997.
Abbreviations: Cl, confidence interval; OR, odds ratio. cedures have been shown to be risk factors that com-
1
Department of General Surgery, University Hospital Utrecht, monly contribute to the formation of inguinal hernia
Utrecht, The Netherlands. (11). However, to our knowledge, all of these risk
2
Department of Epidemiology & Public Health, University of
Utrecht, Utrecht, The Netherlands. factors have never been quantified. Therefore, to study
3
The members of the Coala Trial Group are as follows: W. S. these alleged risk factors, we performed a hospital-
Meijer, St. Clara Hospital, and R. U. Boelhouwer, Ikazia Hospital, based case-control study with incident female cases in
Rotterdam; G. J. Clevers, Diakonessenhuis, and Y. van der Graaf,
M. S. L. Liem, A. J. P. Schrijvers, and Th. J. M. V. van Vroonhoven, six hospitals in the Netherlands.
University Hospital, Utrecht; J. P. Vente, Hofpoort Hospital, Woer-
den; and L. P. S. Stassen, C. J. van Steensel,* and W. F. Weidema,* MATERIALS AND METHODS
Reinier de Graaf Gasthuis, Delft, The Netherlands. (*Presently at the
Ikazia Hospital, Rotterdam, The Netherlands.) Cases and controls
Reprint requests to Dr. M. S. L Liem, Department of General
Surgery, G04.228, University Hospital, P.O. Box 85.500, 3508 GA The trial-bureau of a large multicenter prospective
Utrecht, The Netherlands. randomized trial comparing laparoscopic and conven-
721
722 Liem et al.
tional inguinal hernia repair collected and registered connective tissue disease, and family history for in-
all incident female cases of inguinal hernia in the six guinal hernia, including gender and relation with the
participating hospitals between January 1994 and family member.
November 1995. These hospitals were representative Lifetime physical activity was estimated with two
of the different hospital types in the Netherlands in validated questionnaires with the questions written in
order to enhance generalizability (12). Dutch (14, 15). Physical activity was measured with
An inguinal hernia was diagnosed by two experi- the Baecke questionnaire (14, 16). This questionnaire
enced physicians and was defined as a clinically de- estimates recent, present activity in three categories:
tectable swelling in the groin or a clearly palpable work, sports activity, and leisure time. Additional
defect of the abdominal wall in the groin. For all cases, leisure-time activity questions were added, and the
an operation report was obtained to confirm diagnosis entire questionnaire recently underwent a validation
to exclude uncertainty. Inclusion criteria for cases (16). For each category, high activity was represented
were written informed consent and age between 20 by the maximum score of 5, and low activity by the
and 80 years. Patients who were mentally incompetent minimum score of 1. A total activity index was cal-
or unable to speak and understand Dutch were ex- culated by adding the three separate scores (16). Work
cluded. activity in the past was estimated with four categories
Controls were selected from females who visited the (i.e., sedentary, standing, labor, heavy labor) using
examples for both activity level and job type (17).
matched, unadjusted odds ratios were calculated. A p TABLE 1. Percentage distributions of selected baseline
value of <0.05 was considered statistically significant, characteristics and mean height and weight of cases
and controls in a hospital-based case-control study of
and all p values were calculated two-sided. inguinal hernia in women, the Netherlands, January 1994
Significant univariate correlates found with matched to November 1995
(i.e., conditional) logistic regression were entered into
Cases Controls
a multivariate conditional logistic regression analysis Characteristic (n = 72) (n=125)
to correct for the simultaneous effects of covariates Age category (years) (%)
(EGRET version 1.02.07, 1995, Statistics and Epide- <35 9.7 9.7
miology Research Corp. and Cytel Software Corp., >35-:50 33.3 32.0
Seattle, Washington). In addition, other variables >50-<65 23.6 30.4
>65 33.3 28.0
whose univariate test had a p value of <0.25, or Height in cm (SD*) 167(7) 166 (6)
variables which were of alleged biologic importance, Weight in kg (SD) 67(11) 69(11)
were also included in the model. For model building, Education (%)
we applied backward stepwise elimination of variables None 1.4 1.6
Elementary 40.3 52.8
(20). Adjusted odds ratios with their 95 percent con- Secondary 54.2 42.4
fidence intervals were derived from the estimated re- University 4.2 3.2
gression coefficients. Marital status (%)
No partner 20.8 14.4
TABLE 2. Potential risk factors in cases and controls in a hospital-based case-control study of
inguinal hernia in women, the Netherlands, January 1994 to November 1995
Present activity
Present work activity index 2.9 (2.6-3.2) 2.9(2.6-3.1) 0.6 (NSt)
Duration present work (years) 26 (17-43) 30(12-43) 1.0 (NS)
Duration x present work activity 86.9 (43.6-118. 1) 86.6 (36.3-123.5) 1.0 (NS)
Present sports activity index 2.0(1.8-2.5) 2.3(1.8-2.8) 0.02
Present leisure-time activity index 2.9 (2.7-3.3) 3.0 (2.6-3.3) 1.0 (NS)
Present total index (work + sports + leisure
time) 8.0 (7.2-8.5) 8.1 (7.3-9.0) 0.15 (NS)
Climbing stairs/day (no.) 7(3-11) 10(5-12) 0.03
Past activity
Past work activity index 1(0-3) 2(0-3) 0.9 (NS)
Duration past work (years) 5(0-11) 6 (0-10) 0.5 (NS)
Duration x past work activity 6 (0-24) 12(0-25) 0.4 (NS)
Past sports activity index 6(4-8) 6(4-8) 0.5 (NS)
Past leisure-time activity index 8(6-9) 8(6-9) 0.8 (NS)
23.9 (22.0-25.8) 24.8 (22.8-27.6)
% yes Crude
Risk
95% Clt
factor Cases Controls ratio*
Miscarriage 0 1 0.9 0.4-1.7
Children 81 82 0.9 0.4-1.9
Twins 3 3 0.9 0.2-4.8
Inguinal hernia in family 44 19 3.4 1.8-6.4
Positive female family member 13 2 8.8 1.8-41.9
Smoking 25 26 0.9 0.5-1.8
COPD 6 5 1.2 0.3-4.3
Urinary tract obstruction 7 3 2.3 0.6-8.7
Obstipation 31 18 2.0 1.0-3.8
Trauma 1 2 0.6 0.1-5.6
Pelvic fracture 3 1 3.5 0.3-39.8
Abdominal operations (%) 44 53 0.7 0.4-1.3
Umbilical hernia (%) 4 1 5.4 0.6-52.8
Cicatrical hernia (%) 3 2 1.8 0.2-12.7
Appendectomy (%) 22 22 1.0 0.5-2.1
Cholecystectomy (%) 6 10 0.5 0.2-1.6
Bowel operation for carcinoma (%) 6 6 0.9 0.2-3.0
* p values and odds ratios were calculated by unmatched analyses.
t NS, not significant; Cl, confidence interval; COPD, chronic obstructive pulmonary disease.
variate conditional model. The remaining factors were factors that were found to be independently associated
Quetelet index, obstipation, present habitual sports with inguinal hernia in females were present habitual
activity index, and family history of inguinal hernia, sports activity, Quetelet index, defecation pattern, and
and these variables provided the most stable model family history of inguinal hernia.
(table 3). Adjusted odds ratios controlling for simul- These findings are in contrast to the main risk factor
taneous effects are shown. A high level of present that has been suspected to be associated with inguinal
habitual sports activity was protective for inguinal hernia, namely heavy physical exercise. The stronger
hernia compared with a low level. For instance, fe- inguinal anatomy in females, as shown by a number of
males with a positive family history and high habitual authors (2, 3, 11), may explain this lack of association.
sports activity may not have an increased risk for Heavy physical exercise has been the subject of many
inguinal hernia. statements and studies in the literature (21, 22). Some
studies have classified occupations into categories for
DISCUSSION activity level (21, 23), thereby also assuming similar
This case-control study has studied all potential risk standard activity for people with the same occupation.
factors as reported in the literature. The only risk For instance, all office clerks perform sedentary work.
TABLE 3. Adjusted odds ratios for inguinal hernia in women the absence of sports activities coincided with obesity
by major risk factors in a hospital-based case-control study, in the females of this report (11) and this may explain
the Netherlands, January 1994 to November 1995 the opposite finding. In addition, hernias may not be so
Risk
Adjusted easy to diagnose in obese females. Other factors such
odds 95% Clt
(actor
ratio* as pregnancies and abdominal operations were also
No inguinal hernia in family 1-0*
suspected to be risk factors but did not show any
Inguinal hernia in family 4.3 1.9-9.7 relation. Ponka (11) observed the presence of these
factors more frequently in females with inguinal her-
Present sports activity index nia in his uncontrolled case series but did not provide
score
<1.75 1.0*
a statistical analysis.
1.75-2.0 0.8 0.3-1.9 Individual predisposition to inguinal hernia has been
2.0-2.75 0.4 0.1-1.0 claimed in the past mostly on the basis of the various
>2.75 0.2 0.1-0.7 anatomy found in dissection studies (8, 24, 25), but
Quetelet index (kg/rtf)
<25 1.0*
this individual predisposition has not been quantified.
>25, <30 0.7 0.3-1.5 A positive family history may point toward a relatively
>30 0.2 0.04-1.0 weak anatomy and may form the basis for this predis-
position. Spangen (2) and Herrington (4) pointed out
No obstipation 1.0*
that many symptomatic but nonpalpable hernias in
an apparent protective effect of sports activities. How- 4. Herrington JK. Occult inguinal hernia in the female. Ann Surg
1975;181:481-3.
ever, all cases with knowledge on this risk factor 5. Wantz GE. Abdominal wall hernias. In: Schwartz SI, Shires
stated that they had not changed their sports activity GT, Spencer FC, et al, eds. Principles of surgery. 6th ed. New
pattern because of the hernia, and it is unlikely that York: McGraw-Hill, Inc, 1994:1517-43.
6. Gue S. Development of right inguinal hernia following appen-
this bias could be the sole explanation because most dicectomy, a 10 year review of cases. Br J Surg 1972;59:
patients see a physician very soon after discomfort that 352-3.
limits their activity, and many patients with inguinal 7. Ryan EA. Hernias related to pelvic fractures. Surg Gynecol
hernia have no discomfort or pain at all (4). Obstet 1971;133:441-6.
8. Donahue P. Theoretic aspects of hernia. In: Nyhus LM, Con-
Finally, we examined obesity and socioeconomic don RE, eds. Hernia. Philadelphia: JB Lippincott, 1995:
status by both patient and partner education for con- 73-82.
founding, but this did not change the odds ratios 9. Cannon DJ, Read RC. Metastatic emphysema. A mechanism
for acquiring inguinal herniation. Ann Surg 1981;194:270-8.
significantly. In addition, matching for age was suc- 10. Peacock EE, Madden JW. Studies on the biology and treat-
cessful. The last concern may be absence of statistical ment of recurrent inguinal hernia. II. Morphological changes.
power. However, based on available data, we assumed Ann Surg 1974; 179:567-71.
11. Ponka JL. Hernias of the abdominal wall. Philadelphia: WB
a standard deviation of 1.1 on the total activity index Saunders, 1980:82-90.
score (16), and we calculated that with our sample size 12. Liem MSL, Steensel CJ van, Boelhouwer RU, et al. The
a difference of 0.66 between the mean total activity learning curve for totally extraperitoneal laparoscopic inguinal
hernia repair. Am J Surg 1996; 171:281-5.