World J. Surg. 26, 397– 400, 2002 DOI: 10.1007/s00268-001-0238-6

WORLD Journal of

SURGERY © 2002 by the Socie´te´ Internationale de Chirurgie

Original Scientific Reports Smoking Is a Risk Factor for Recurrence of Groin Hernia Lars Tue Sorensen, M.D.,1 Esbern Friis, M.D.,1 Torben Jorgensen, M.D., D.M.Sci.,2 Bo Vennits, M.D.,1 Betina Ristorp Andersen, M.D.,1 Gitte Iben Rasmussen, M.D.,1 Johan Kjaergaard, M.D., D.M.Sci.1 1

Department of Surgical Gastroenterology K, Bispebjerg University Hospital, Copenhagen Hospital Corporation, Bispebjerg Bakke 23, DK-2400 Copenhagen NV, Denmark Centre of Preventive Medicine, Medical Department M, Glostrup University Hospital, Copenhagen County, Entrance 8, 7th Floor, DK-2600 Glostrup, Denmark

2

Published Online: January 2, 2002 Abstract. Studies of connective tissue from patients with inguinal hernia have shown that smoking may be associated with hernia formation due to a defective connective tissue metabolism. Whether smoking is a risk factor for recurrence, too, was examined in this study. From December 1990 through December 1995, 649 patients underwent hernia repair as open sutured repair (Cooper ligament or abdominal ring repair) or as open mesh repair. Five hundred forty-four eligible patients were evaluated for recurrence 2 years postoperatively. Association between recurrence and 17 patient-, disease-, and intraoperative variables were analyzed by multiple logistic regression. The results showed that smoking was significantly and independently associated with recurrence compared to nonsmoking [odds ratio (OR ⴝ 2.22; 95% confidence interval (95% CI) ⴝ 1.19 – 4.15)]. Open sutured repair compared to open mesh repair was the most significant predictor for recurrence (OR ⴝ 7.23; 95% CI ⴝ 3.01– 17.37). Surprisingly, local anesthesia was associated with a higher risk of recurrence compared to general anesthesia (OR ⴝ 2.44; 95% CI ⴝ 1.19 –5.09). Potential confounders and other risk factors for hernia recurrence such as age, alcohol consumption, previous surgery, and anatomical characteristics of the hernia were adjusted for in the analysis. In conclusion, smoking is an important risk factor for recurrence of groin hernia, presumably due to an abnormal connective tissue metabolism in smokers.

The introduction of open mesh repair in inguinal hernia repair has proven beneficial for the patients in terms of less postoperative pain, earlier return to work, and lower risk of recurrence [1]. Although the conventional methods of open hernia repair have been superseded, recurrences still occur after mesh repair. It is a general assumption that defective surgical technique is the prime cause for recurrence [2], but risk factors for recurrence other than surgical technique are poorly recognized. Defective connective tissue metabolism in smokers has been suggested as a causative factor for hernia formation [3], but it remains unknown whether smoking is associated with hernia recurrence. The aim of this study was to examine whether smoking is a possible risk factor for recurrence when adjusting for other risk factors and potential confounders related to patient history, Correspondence to: L.T. Sorensen, M.D.

hernia disease, and operation through multiple logistic regression analysis. Materials and Methods From December 1990 through December 1995, two cohorts of a total of 649 eligible patients underwent inguinal hernia repair in one general surgical department. The methods of repair were open sutured repair (Cooper ligament repair or abdominal ring repair) or open mesh repair. The first cohort comprised 188 patients who had undergone hernia repair until March 1993 (Table 1). All patients had been selected for inclusion in a randomized clinical trial comparing the recurrence rate 2 years after open mesh repair versus open sutured repair [4]. The second cohort comprised 461 consecutive patients who underwent hernia repair with identical techniques in the subsequent period (Table 1). Patients who had died or were lost to follow-up were excluded. On follow-up 2 years after surgery, 472 patients were examined for recurrence by a surgeon. Seventy-two patients from the consecutive cohort who failed to appear at examination completed a questionnaire sent by mail. The open mesh repair was introduced in the department in 1990. The procedure was performed as described by Lichtenstein et al. [5] using a 60 ⫻ 110 mm polypropylene mesh (Prolene, Ethicon, Norderstedt, Germany) and a continuous 0-0 prolene suture to retain the mesh in the desired position. The Cooper ligament and abdominal ring repair were both performed as described by McVay [6, 7] using 0 polyglycolic acid absorbable suture (Vicryl, Ethicon, Norderstedt, Germany) to reconstruct the posterior inguinal wall. All members of the staff operated on the patients (consultants, senior registrars with surgical experience of 10 years or more, and trainees with 3– 6 years surgical experience). Being a standard surgical procedure, all surgeons were familiar with the Cooper ligament and abdominal ring repair. Surgeons unfamiliar with the mesh-implantation technique were trained and supervised by experienced colleagues until they felt confident with the method.

398

World J. Surg. Vol. 26, No. 4, April 2002

Table 1. Source of patient material.

Patients in study Patients dead prior to examination Patients lost to follow-up Patients evaluated 2 years postoperatively

Selected cohorta

Consecutive cohortb

Pooled cohort

188 8 1 179

461 54 42 365c

649 62 43 544

a

Herniotomies performed from December 1990 through March 1993. Herniotomies performed from April 1993 through December 1995. c Questionnaires from 7 out of 72 patients indicated hernia recurrence. When the questionnaires were received, these patients were contacted by telephone and 2 patients complied with the definition of hernia recurrence (reoperation or provision with a truss) [9]. b

Primary hernia repairs were performed as day case surgery with the patients leaving the hospital 2– 8 hours after surgery unless a complication or concomitant disease indicated postoperative admission. Primary hernia repairs were performed under local anesthesia as described by Amid et al. [8], if the patient accepted; all other operations were performed under general or spinal anesthesia. Antibiotics were not used routinely. Patients were given no restrictions in postoperative mobility and were recommended to return to work after 1–2 weeks (heavy workers after 4 weeks). Data regarding recurrence were collected at examination by surgeons in the outpatient clinic 2 years postoperatively or by questionnaire sent to patients who failed to appear. Hernia recurrence was considered to be present if a bulge or weakness in the operation area exacerbated by a Valsalva maneuver necessitated a further operation or provision of a truss [9]. Sixteen different variables related to patient history, hernia disease, and operation with possible relation to recurrence were assessed. In addition, a variable characterizing each cohort was made in order to adjust for possible selection bias in the material and thus permit pooling of data. Data regarding patient history (age, gender, body mass index, chronic obstructive pulmonary disease, occupational status, smoking and drinking habits) were collected from self-administered questionnaires completed prospectively by elective patients at referral to the outpatient clinic prior to surgery. Patients operated on an emergency basis were interviewed at the time of admission by a registrar. The patients reported their smoking habits as number of cigarettes, cheroots, cigars, or pipe fills per day and their consumption of alcohol as number of drinks per week. One drink was defined as containing 9 –13 g of alcohol being equivalent to one bottled beer, one glass of wine, or one measure of spirits. Other data were collected from clinical records (emergency or elective admission, postoperative admission time, postoperative complications), which also contained a description of surgical procedure (name of surgeon, anatomical characteristics of the hernia, side, previous surgery, anesthesia, and method of repair). The results were analyzed by multiple logistic regression (SPSS for Windows 8.0, SPSS Inc., Chicago, Illinois, USA). Hernia recurrence 2 years after surgery was the dependent variable. First, a univariate analysis was performed with patient age and cohort as fixed covariates. Based on this model, the odds ratio of each variable was estimated. Second, variables likely associated with recurrence (p ⱕ 0.2) were included in a final multivariate model of logistic regression. Finally, interaction terms between variables

Table 2. Variables associated with hernia recurrence analyzed by multiple logistic regression: the final model. Variable Age Continuous variable Cohort Selected cohort Consecutive cohort Smoking habits Nonsmoker Smoker Missing data Alcohol consumption Abstainer 1–7 drinks per week 8 –14 drinks per week 15–21 drinks per week ⬎21 drinks per week Missing data Anesthesia General Local Regional Anatomical characteristics Indirect hernia Direct herniaa Previous surgery Primary hernia Recurrent hernia Method of hernia repair Open mesh repair Open sutured repairb

n

Univariate odds ratio

Multivariate odds ratio

544

1.02 (1.00 –1.04)

1.01 (0.99 –1.03)

179 365

1.00 (–) 0.61 (0.36 –1.04)

1.00 (–) 0.62 (0.32–1.22)

265 255 24

1.00 (–) 1.51 (0.88 –2.56)

1.00 (–) 2.22 (1.19 – 4.15)

250 71 65 49 40 69

1.00 (–) 0.46 (0.17–1.22) 0.92 (0.42–2.02) 0.51 (0.17–1.50) 1.00 (0.39 –2.59)

1.00 (–) 0.30 (0.10 – 0.93) 0.97 (0.40 –2.34) 0.37 (0.11–1.16) 0.91 (0.33–2.55)

249 242 53

1.00 (–) 1.39 (0.79 –2.47) 1.71 (0.75–3.88)

1.00 (–) 2.44 (1.19 –5.09) 1.70 (0.61– 4.78)

240 304

1.00 (–) 2.76 (1.53– 4.99)

1.00 (–) 5.00 (2.50 –10.02)

437 107

1.00 (–) 1.76 (0.99 –3.13)

1.00 (–) 3.56 (1.59 –7.98)

161 383

1.00 (–) 3.89 (1.84 – 8.23)

1.00 (–) 7.23 (3.01–17.37)

Values in parentheses are 95% confidence intervals. Number of patients included in the analysis: 475 (minus cases with missing data). a Includes femoral and combined hernia. b Cooper ligament or abdominal ring repair.

were examined. All results were described with odds ratio and 95% confidence interval. Results In the pooled cohort, the recurrence rate after primary repair was 11.2% (8.4 –14.6) and 18.7% (11.8 –27.4) after secondary repair. Median age was 63 years (range 15–90 years). The univariate analysis revealed three variables significantly associated with recurrence: age, anatomical characteristics of the hernia, and method of repair (Table 2). The association between smoking habits, alcohol consumption, previous surgery, and anesthesia with recurrence showed a nonsignificant trend (p ⱕ 0.2), whereas the other variables, listed in Table 3, were not associated with recurrence. The multivariate logistic regression analysis disclosed that smoking compared to nonsmoking was significantly and independently associated with recurrence of groin hernia (Table 2). Alcohol consumption of 1–7 drinks per week was negatively associated with recurrence as compared with abstention. Patients operated with local anesthesia had a higher risk of recurrence compared to patients operated with general anesthesia. Open sutured repair was associated with a higher risk of recurrence compared to open mesh repair. Finally, patients with recurrent as well as direct hernia had a higher risk of recurrence compared to patients with primary or indirect hernias.

Sorensen et al.: Smoking and Groin Hernia

399

Table 3. Variables with no significant relation to hernia recurrence in 544 patients. Variable Gender (female sex) Mean (S.D.) body mass index Chronic obstructive pulmonary disease Employment status Employed patients Unemployed patients (including retirees) Employment status unknown Side (right sided) Stage of training of surgeon Herniotomies performed by specialized surgeonb Herniotomies performed by trainees Postoperative complications Wound infection Hematoma Emergency operation Median (range) postoperative admission time

na 55 24.3 36

% (10.1) (1.94) (6.6)

241 272 31 334

(44.3) (50.0) (5.7) (61.4)

230 314

(42.3) (57.7)

13 38 24 1 day

(2.4) (7.0) (4.4) (1–35)

a Values are number of patients (with percentages in parentheses) unless stated otherwise. b Supervised operations are credited to the specialist surgeon.

No dose-response association between smoking and recurrence of groin hernia was found. Likewise, no significant interaction between smoking and alcohol consumption in relation to recurrence was detected. When adjusting for cohort, patient age, smoking habits, alcohol consumption, anesthesia, previous surgery, anatomical characteristics of the hernia, and method of repair, no significant association was established between trainees compared to specialized surgeons and recurrence (OR ⫽ 1.28; 95% CI ⫽ 0.69 –2.40). By forward logistic regression the association between smoking and recurrence became significant when alcohol consumption was included in the final model. Alcohol consumption became significant when smoking, previous surgery, and either anatomical characteristics of the hernia or anesthesia were included in the model. Local anesthesia turned significant when method of repair, previous surgery, and anatomical characteristics of the hernia were included in the model. Finally, previous surgery turned significant when either smoking habits, alcohol consumption, method of repair, or anesthesia were included in the model. Age lost its significant association with recurrence when anatomical characteristics of the hernia or alcohol consumption were included in the model. Discussion This study demonstrates that smoking is significantly associated with recurrence of groin hernia independent of patient age, alcohol consumption, previous hernia surgery, anesthesia, anatomical characteristics of the hernia, and method of repair. The overall recurrence rate in this study of 12.7% (10.0 –15.8) is comparable with reports from other nonspecialized centers [10]. As in other studies [11], patient age was not an independent risk factor for recurrence. Likewise, chronic obstructive pulmonary disease indicating extensive coughing and increased intraabdominal pressure was not associated with recurrence. The cohort variable revealed that patients operated in the consecutive series were less likely to have a recurrence than patients operated in the selected series. Although not statistically significant, this result

may reflect an underestimation of the recurrence rate in the consecutive series, as 20% (72/365) of these patients were evaluated by questionnaire and not by examination by a surgeon [12]. Previous reports have demonstrated that smokers have a higher risk of postoperative complications after surgery than nonsmokers [13–15]. Following this study, recurrence 2 years after hernia repair can be added to the list. In patients with inguinal herniation but with unknown smoking habits, impaired collagen synthesis and connective tissue pathology have been identified [16 –18]. Recent evidence has demonstrated that smokers have a diminished collagen deposition possibly leading to a reduced tensile strength in wounds [19]. This may be due to impaired cellular and collagen biosynthesis as a result of smoking-induced tissue hypoxia [20, 21]. Another hypothesis suggests excessive proteolysis in smokers as a result of stimulated neutrophil-macrophage response and a high serum elastolytic activity. This systemic protease/antiprotease imbalance impairs wound healing and damages collagen and connective tissue [3]. The relatively high proportion of smokers in this study as compared to the Danish population in general (46.8% versus 37.0% [1993]) [22] suggests like others, that the prevalence of inguinal herniation is higher in smokers than in nonsmokers [23]. The percentage of patients with previous hernia surgery who smoked was 45%. This finding illustrates that risk factors other than smoking play a significant role in the pathogenesis of hernia recurrence and must therefore be adjusted for. The association between alcohol consumption and recurrence shows that drinking between 1 and 7 drinks per week was negatively associated with recurrence. It has previously been reported that patients with a habitual intake of more than 5 drinks per day have a higher risk of postoperative complications [13, 15, 24]. If hernia recurrence is considered as a late complication, this study cannot confirm these results. However, a possible beneficial effect of moderate alcohol intake is in accordance with other reports regarding the association of alcohol consumption and postoperative morbidity [13, 25]. Although the effect of anesthesia on recurrence after hernia repair has not been studied, it is generally assumed that the type of anesthesia does not influence recurrence rate [26]. In our study, local anesthesia was coincidentally found to be associated with hernia recurrence. In contrast to a previous report [27], this finding was independent of the surgeons’ stage of training. A possible explanation is incomplete anesthetic block achieved by local anesthesia [28], as patient discomfort and reduced cooperation may have affected the surgeon’s technical performance resulting in insufficient hernia repair. As regards the significance of the method of repair, only a few studies have compared recurrence after open mesh repair and open sutured repair [1]. However, our findings showing a reduced recurrence rate after open mesh repair compared to open sutured repair confirm results previously reported from our group and others [4, 29]. In addition, the study confirms a recent report showing that risk of recurrence in both direct and recurrent hernias is significantly higher than in primary or indirect hernia [10]. In conclusion, the risk of recurrence of groin hernia is significantly increased in smokers, presumably due to an abnormal connective tissue metabolism. This finding may imply that surgeons would be justified to advise patients to quit smoking before undergoing hernia repair. However, further research is needed to

400

determine whether the detrimental effect of smoking on connective tissue healing is reversible through abstinence prior to surgery. Résumé. Les études de tissu conjonctif provenant des patients ayant une hernie inguinale ont montré que le fait de fumer pourrait influencer la survenu de hernie en raison d’un défaut de métabolisme du tissu collagène. Dans cette étude, nous avons examiné si le fait de fumer était un facteur de risque également de récidive. Entre décembre 1990 et décembre 1995, on a réparé 649 hernies par une raphie (technique de McVay ou annuloplastie) ou par prothèse par voie ouverte. Cinq cents quarante-quatre patients ont été évalués pour récidive deux ans après opération. L’association entre la récidive et 17 paramètres concernant le patient, la maladie ou les facteurs peropératoires ont été examinés par la méthode d’analyse de régression logistique. Les résultats montrent que le fait de fumer était significativement et indépendamment associé à la récidive comparé aux patients qui ne fumaient pas [“odds ratio” (OR ⴝ 2.22; intervalle de confiance à 95% (CI 95%) ⴝ 1.19 – 4.15)]. La méthode de raphie par voie ouverte, comparé à la réparation par prothèse, a été le facteur de risque de récidive le plus hautement significatif (OR ⴝ 7.23; CI 95% ⴝ 3.01–17.37). De façon surprenante, l’anesthésie locale a été associée avec un risque plus élevé de récidive comparé à l’anesthésie générale (OR ⴝ 2.44; CI 95% ⴝ 1.19 –5.09). Ces résultats restaient similaires, même après ajustement pour les autres facteurs de risque potentiels de récidive de hernie tels que l’âge, la consommation d’alcool, un antécédent de chirurgie et les caractéristiques de la hernie. En conclusion, le fait de fumer est un facteur de risque important dans la récidive de hernie inguinale: on présume que ce risque est en rapport avec un métabolisme de tissu conjonctif perturbé chez le fumeur. Resumen. Diversos estudios con tejido conjuntivo de pacientes con hernia inguinal han demostrado, que fumar constituye un factor propiciatorio en la génesis de la hernia, por su acción deletérea sobre el metabolismo del tejido conectivo. En este estudio intentamos averiguar si fumar puede también ser un factor de riesgo en la recidiva herniaria. Desde diciembre de 1990 al mismo mes de 1995, 649 pacientes fueron sometidos a una reparación herniaria mediante cirugía abierta (herniorrafia al ligamento de Cooper, reparación del anillo inguinal interno) o con hernioplastia con mallas. 544 pacientes fueron reexaminados a los dos años de la intervención con objeto de detectar recidivas herniarias. Mediante una regresión logística múltiple se analizó la asociación entre la recidiva y 17 enfermedades de los pacientes así como, con diversas variables intraoperatorias. Los resultados demostraron que fumar (al comparase con los no fumadores), es un factor significativo e independiente que se asocia a la recidiva herniaria [“odds ratio” (OR ⴝ 2.22; 95% de intérvalo de confianza) (95% CI) ⴝ 1.19 – 4.15]. La herniorrafia convencional en comparación con la hernioplastia con malla constituye el factor más importante por lo que a la recidiva herniaria se refiere [OR ⴝ 7.23; 95 CI ⴝ 3.01–17.37]. Curiosamente, observamos que la anestesia local se acompaña de un mayor riesgo de recidiva que la detectada en pacientes intervenidos bajo anestesia general [OR ⴝ 2.44; 95 CI ⴝ 1.19 –5.09]. Se confirmó que otros factores conocidos tales como: la edad, consumo de alcohol, cirugía previa y características anatómicas de la hernia, también facilitan la recidiva herniaria. En conclusión, fumar constituye un importante factor favorecedor de la recidiva herniaria debido, probablemente, al metabolismo anormal del tejido conjuntivo de los fumadores.

Acknowledgments The authors thank nurse Mette Jensen for her assistance in locating and retrieving the case records of the patients in this study. The support of Danish Research Council (case no. 9801273) is also acknowledged. References 1. Cheek C, Black NA, Devlin HB, et al. Groin hernia surgery: a systematic review. Ann. R. Coll. Surg. 1998;80:S1–S19

World J. Surg. Vol. 26, No. 4, April 2002 2. Lichtenstein IL, Shulman AG, Amid PK. The cause, prevention, and treatment of recurrent groin hernia. Surg. Clin. North Am. 1993;73: 529 –544 3. Cannon DJ, Read RC. Metastatic emphysema: a mechanism for acquiring inguinal herniation. Ann. Surg. 1981;194:270 –278 4. Friis E, Lindahl F. The tension-free hernioplasty in a randomized trial. Am. J. Surg. 1996;172:315–319 5. Lichtenstein IL, Shulman AG, Amid PK, et al. The tension-free hernioplasty. Am. J. Surg. 1989;157:188 –193 6. McVay CB. The Pathologic Anatomy of the More Common Hernias and Their Anatomic Repair. Springfield, Illinois, Charles C. Thomas Publisher, 1954;15–28 7. Halverson K, McVay CB. Inguinal and femoral hernioplasty. Arch. Surg. 1970;101:127–135 8. Amid PK, Shulman AG, Lichtenstein IL. Open “tension-free” repair of inguinal hernias: the Lichtenstein technique. Eur. J. Surg. 1996; 162:447– 453 9. Marsden AJ. The results of inguinal hernia repairs. Lancet 1959;1: 461– 462 10. Kald A, Nilsson E, Anderberg B, et al. Reoperation as surrogate endpoint in hernia surgery. A three year follow-up of 1565 herniorrhaphies. Eur. J. Surg. 1998;164:45–50 11. Nilsson E, Kald A, Anderberg B, et al. Hernia surgery in a defined population: a prospective three year audit. Eur. J. Surg. 1997;163:823– 829 12. Vos PM, Simons MP, Luitse JSK, et al. Follow-up after inguinal hernia repair. Questionnaire compared with physical examination: a prospective study in 299 patients. Eur. J. Surg. 1998;164:533–536 13. Sorensen LT, Jorgensen T, Kirkeby LT, et al. Smoking and alcohol abuse are major risk factors for anastomotic leakage in colorectal surgery. Br. J. Surg. 1999;86:927–931 14. Horby J, Jorgensen T, Piilsgaard B. Risk factors and their importance in complications after breast surgery. Eur. J. Cancer 1996;32A:159 15. Stopinski J, Staib I, Weissbach M. Do abuse of nicotine and alcohol have an effect on the incidence of postoperative bacterial infections? J. Chir. 1993;130:422– 425 16. Wagh PV, Read RC. Defective collagen synthesis in inguinal herniation. Am. J. Surg. 1972;124:819 – 822 17. Ajabnoor MA, Mokhtar AM, Rafee AA, et al. Defective collagen metabolism in Saudi patients with hernia. Ann. Clin. Biochem. 1992; 29:430 – 436 18. Pans A, Pierard GE, Albert A, et al. Adult groin hernias: new insight into their biomechanical characteristics. Eur. J. Clin. Invest. 1997;27: 863– 868 19. Jorgensen LN, Kallehave F, Christensen E, et al. Less collagen production in smokers. Surgery 1998;123:450 – 455 20. Jensen JA, Goodson WH, Hopf HW, et al. Cigarette smoking decreases tissue oxygen. Arch. Surg. 1991;126:1131–1134 21. Jonsson K, Jensen JA, Goodson WH III, et al. Tissue oxygenation, anemia, and perfusion in relation to wound healing in surgical patients. Ann. Surg. 1991;214:605– 613 22. Ramstrom L. Prevalence and other dimensions of smoking in the world. In Bolliger CT, Fagerstrom KO. The Tobacco Epidemic. Prog. Respir. Res. 1997;28:64 –77 23. Bielecki K, Pulawski R. Is cigarette smoking a causative factor in the development of inguinal hernia? Pol. Tyg. Lek. 1988;43:974 –976 24. Tonnesen H, Kehlet H. Preoperative alcoholism and postoperative morbidity. Br. J. Surg. 1999;86:869 – 874 25. Espehaug B, Havelin LI, Engesaeter LB, et al. Patient-related risk factors for early revision of total hip replacements. A population register-based case-control study of 674 revised hips. Acta Orthop. Scand. 1997;68:207–215 26. Abrahamson J. Etiology and pathophysiology of primary and recurrent groin hernia formation. Surg. Clin. North Am. 1998;78:953–972 27. Kingsnorth AN, Britton BJ, Morris PJ. Recurrent inguinal hernia after local anaesthetic repair. Br. J. Surg. 1981;68:273–275 28. Rutkow IM, Robbins AW. “Tension-free” inguinal herniorrhaphy: a preliminary report on the “mesh plug” technique. Surgery 1993;114: 3– 8 29. Janu PG, Sellers KD, Mangiante EC. Mesh inguinal herniorrhaphy: a ten year review. Am. Surg. 1997;63:1065–1071