- Email: [email protected]
Risk factors for postoperative hematoma after inguinal hernia repair: an update
j o u r n a l o f s u r g i c a l r e s e a r c h s e p t e m b e r 2 0 1 6 ( 2 0 5 ) 3 3 e3 7
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Risk factors for postoperative hematoma after inguinal hernia repair: an update Muhammad H. Zeb, MBBS, T.K. Pandian, MD, Moustafa M. El Khatib, MBBCh, Nimesh D. Naik, MD, Abhishek Chandra, David S. Morris, MD, Rory L. Smoot, MD, and David R. Farley, MD* Division of Subspecialty General Surgery, Mayo Clinic College of Medicine, Rochester, Minnesota
article info
abstract
Article history:
Background: We recently sensed an increase in the frequency of groin hematoma after
Received 12 February 2016
inguinal hernia repair (IHR) at our institution. The aim of this study was to provide a more
Received in revised form
updated assessment of the risk factors inherent to this complication.
19 May 2016
Methods: We performed a case-control study of all adult patients (age 18 y) who developed
Accepted 1 June 2016
a groin hematoma after IHR at our institution between 2003 and 2015. Univariate and
Available online 8 June 2016
multivariable analyses were performed to assess for independent predictors for groin hematoma.
Keywords:
Results: A total of 96 patients (among 6608 IHR) developed a groin hematoma, (60 were
Inguinal hernia
observed, 36 required intervention). The hematoma frequency increased from our previous
Inguinal herniorrhaphy
study (1.4 % versus 0.9%, P < 0.01). Mean age was 64.6 y (range: 18-92), and 84.3% were men.
Postoperative hematoma
There was no significant difference in the laterality, type, or technique of IHR between
Groin hematoma
cases and controls. Univariate analysis (odds ratio [95% confidence interval], P) identified
Anticoagulation
warfarin usage (3.5, [1.6-6.4], P < 0.01), valvular heart disease (11.6, [2.6-51.3], P < 0.01), atrial fibrillation (2.6, [1.2-5.5], P ¼ 0.01), hypertension (2.03, [1.1-3.6], P ¼ 0.02), recurrent hernia (3.7, [1.4-9.7], P < 0.01), and coronary artery disease (2.1, [1.0-4.4 ], P ¼ 0.05) as significant preoperative factors. The proportion of patients on warfarin decreased since our prior report (31% versus 42%, P ¼ 0.20). On multivariable regression, warfarin and recurrent hernia were independent predictors of hematoma development. Conclusions: Independent risk factors for the development of groin hematoma after IHR included warfarin use and recurrent hernia. Careful consideration for anticoagulation and surgical hypervigilance remains prudent in all patients undergoing IHR and especially those with recurrence. ª 2016 Published by Elsevier Inc.
Introduction Inguinal hernia repair (IHR) is one of the most common surgical procedures worldwide. In the United States alone, approximately 800,000 repairs are performed annually.1 Groin hematoma after IHR is an infrequent complication that can
cause significant patient discomfort, require reoperation, and delay postoperative recovery. Several studies reveal that postoperative groin hematomas after IHR occur with a frequency between 0.3% and 6%.2-8 We have previously reported an increase in hematoma occurrence in patients who were on warfarin therapy undergoing IHR.7 A concern has always
* Corresponding author. Department of Surgery, Mayo Clinic, 200 First Street, SW, Rochester, MN 55905. Tel.: þ1 507 284 2095; fax: þ1 507 284 5196. E-mail address: [email protected] (D.R. Farley). 0022-4804/$ e see front matter ª 2016 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jss.2016.06.002
34
j o u r n a l o f s u r g i c a l r e s e a r c h s e p t e m b e r 2 0 1 6 ( 2 0 5 ) 3 3 e3 7
existed regarding the risk of bleeding associated with cessation and resumption of anticoagulation after surgery. Earlier studies have focused on anticoagulation as an important predictor of groin hematoma but have not provided much detail regarding optimal bridging therapy.8-10 Whether bridging with heparin is a contributor to hematoma remains unclear, but a recent randomized study has suggested that bridging triples the risk of bleeding.11 In addition to established coagulation abnormalities, other possible contributors to hematoma formation may include hernia location, recurrence, or incarceration.12 Laparoscopic hernia repair has been proven to have faster recovery and earlier return to work compared with Lichtenstein repair and is the treatment of choice in many patients.13,14 As older surgeons adopt the laparoscopic approach, rates of hematoma may increase as they work upward along the necessary learning curve.15 Herein, we aim to review our experience with groin hematoma after IHR, update our results as previously published, and further analyze risk factors unrelated to anticoagulation.
Methods
with heparin, and anticoagulation restart time frame. In addition, patient medical conditions including presence of cardiac valvular disease, atrial fibrillation, cerebral vascular disease, hypertension, hyperlipidemia, coronary artery disease, pulmonary embolism, hematologic abnormalities (leukemia, polycythemia, and thrombocytopenia) and evidence of previous bleeding were evaluated.
Statistical analysis Univariate analysis was initially used to identify patient and operative characteristic differences between cases and controls. To analyze matched data with a binary outcome, a conditional logistic regression model was used. This model was created so that the proportional hazards for each case and associated matched control were in a separate stratum. The odds ratios (ORs) from these models were reported with 95% confidence intervals (CIs). Multivariable models were then built using stepwise, backward, and best subset selection. All tests were two sided with a significance level of 0.05. All analyses were conducted with SPSS (Statistics for Windows, IBM Corp, Released 2011, version 20.0, Armonk, NY).
Cases and controls
Results Under institutional review board approval, all adults (age 18 y) at our institution who developed a groin hematoma after IHR between the years 2003 and 2015 were studied. These patients were identified using the morbidity and mortality lists for all general surgeons operating at our institution. Morbidity and mortality lists at our institution are generated by each surgical team as complications occur. Despite this “rolling” submission, it is certainly possible that subclinical hematomas not requiring any diagnostic modality or intervention may not have been included in the list. Cases were defined as the patients who developed groin hematoma, confirmed by computed tomography scan or ultrasound, within 30 d of the procedure. To avoid overestimation of hematoma occurrence, patients with clinical findings such as ecchymoses and a mass were not included unless imaging confirmed the diagnosis. Controls were obtained by using surgical indexing to identify all patients who had undergone IHR at our institution during the study years and crossreferencing this with the morbidity and mortality lists to remove those that developed hematoma. Thereafter, each case was matched by age (at the time of repair) and gender in 1:1 fashion. Time of operation for matched cases and controls was within 1 y or less to attempt to match for the period of presentation. This sample size provided approximately 80% power to detect a relative risk of 3.0.
Data extraction Both patient (cases and controls) and procedure characteristics were obtained including age, gender, date of repair, type of hernia (direct, indirect or both), defect laterality, hernia recurrence, incarceration, type of repair performed, date of hematoma formation, concomitant medications (warfarin, heparin, aspirin, clopidogrel), use of perioperative bridging
A total of 96 cases (1.4%) of postoperative hematoma after IHR were identified from 6608 hernia repair operations (Table 1). The mean age for cases and controls was 64.6 y (17.1), and 84.3% were men. There was no significant difference between the cases and controls in terms of hernia laterality (left, right, bilateral; P ¼ 0.85, usage of aspirin [P ¼ 0.15], usage of clopidogrel [P ¼ 1.00], and hyperlipidemia [P ¼ 0.19]). Furthermore, both groups were comparable in regard to hernia repair technique (P ¼ 0.58; Table 2). The mean time to hematoma formation and/or recognition was postoperative day (POD) 4.7 (3.8).
Table 1 e Number of hematomas and hernias performed per year 2003-2015. Year
Hernia repairs
Hematomas
%
2003
566
5
0.8
2004
598
9
1.5
2005
542
10
1.8
2006
506
8
1.5
2007
478
5
0.8
2008
564
6
1.0
2009
588
3
0.5
2010
549
7
1.2
2011
565
7
1.2
2012
472
11
2.3
2013
597
7
1.1
2014
523
12
2.2
2015
174
5
2.8
6608
96
1.4
Total
zeb et al hematoma after inguinal hernia repair
Table 2 e Patient characteristics. Variable
Cases (96)
Controls (96)
64 (18-92)
64 (18-92)
n (%)
n (%)
Male
81 (84)
81 (84)
Female
15 (16)
15 (16)
Average age (range)
Gender
Left
30 (31)
30 (31)
Right
47 (49)
44 (49)
Bilateral
19 (19)
22 (19)
Direct
19 (20)
30 (31)
Indirect
46 (48)
44 (46)
9 (9)
14 (15)
Recurrent hernia
19 (20)
6 (6)
Incarcerated hernia
10 (10)
1 (1)
Type of hernia
Type of repair Bassini
2 (2)
1 (1)
Lichtenstein
57 (59)
58 (60)
Mesh and/or plug
14 (14)
9 (9)
Laparoscopic
23 (24)
28 (29)
Warfarin
30 (31)
11 (12)
Heparin
3 (1)
1 (1)
Aspirin
24 (25)
33 (35)
2 (2)
1 (1)
18 (19)
3 (3)
Valvular disease
19 (19)
2 (2)
Atrial fibrillation
26 (27)
12 (13)
Hypertension
51 (53)
35 (36)
Hyperlipidemia
49 (51)
39 (41)
CVD
14 (15)
4 (4)
Diabetes mellitus
10 (10)
10 (10)
Hematologic abnormalities
26 (27)
14 (14)
Previous bleeding
20 (21)
4 (4)
CHF
10 (10)
5 (5)
Medication
Plavix Bridging with heparin
hypertension (OR 2.0, 95% CI [1.1-3.6]; Table 3). Although not statistically significant, the proportion of patients with warfarin usage decreased since our prior report (31% versus 42%, P ¼ 0.20).7 On multivariable logistic regression, warfarin usage, previous bleeding, incarcerated hernia, bridging with heparin, and recurrent hernia were independent predictors for the development of groin hematoma (P < 0.001).
Anticoagulation
Location of hernia
Both
35
Medical conditions
A total of 41 patients (11 [12%] in the control group and 30 [31%] in the case group) were receiving chronic anticoagulation with warfarin at the time of surgery. Of these 41 patients on warfarin, 21 received perioperative heparin bridging. Only three patients in the control group underwent heparin bridging (with lowemolecular weight heparin). Warfarin therapy was reinstituted on average, on POD 1 (range: 0-4). Preoperative international normalized ratio (INR) values were available for 30 of the 96 cases. Twelve of these 30 cases had an INR value 1.3 preoperatively; the remaining 18 patients had values 1.2. Preoperative INR values were available for 10 of the control patients; seven had an INR 1.3. In our cohort, in elective surgeries, warfarin was stopped 5 days before surgery in 90% of hematoma patients; warfarin was stopped 5 days before surgery in all (100%) control group patients. None of the hematoma patients were on newer anticoagulants (direct thrombin inhibitors such as dabigatran, apixaban, and so forth); however, one patient in the control group was on dabigatran. Due to this small number, these anticoagulants were not included in the univariate or multivariable models.
Cases undergoing evacuation
CHF ¼ congestive heart failure; CVD ¼ cardiovascular disease.
On univariate analysis, history of cardiac valvular disease (P < 0.01), incarcerated hernia (P ¼ 0.01), bridging with heparin (P < 0.01), previous bleeding (P < 0.01), recurrent hernia (P < 0.01), perioperative warfarin requirement (P < 0.01), atrial fibrillation (P ¼ 0.01), coronary artery disease (P ¼ 0.05), and hypertension (P ¼ 0.02) were identified as risk factors for hematoma formation. Cardiac valvular disease had the highest estimated OR (11.6, 95% CI [2.6-51.3]), followed by incarcerated hernia (OR 11.0, 95% CI [1.3-88]), bridging with heparin (OR 7.1, 95% CI [2.0-25.1]), previous bleeding (OR 6.1, 95% CI [1.9-18.4]), recurrent hernia (OR 3.7, 95% CI [1.4-9.7]), warfarin (OR 3.5, 95% CI [1.6-6.4]), atrial fibrillation (OR 2.6, 95% CI [1.2-5.5]), coronary artery disease (OR 2.1, 95% CI [1.0-4.4]), and
Thirty-six patients (38%) had a postoperative hematoma requiring evacuation. Of these, 30 patients underwent reoperation. An active bleeding vessel was identified at the time of evacuation in seven patients. No active bleeding was noted in an additional 20 patients, whereas the remaining three patients were noted to have generalized “oozing” at the time of evacuation. Six patients only underwent needle aspiration of their hematomas.
Morbidity and mortality One patient developed a groin hematoma on POD 6 and was taken to the operating room that day for operative evacuation. Unfortunately, this patient suffered massive blood loss and died on POD 17 due to multiorgan system failure from hemorrhagic shock. Six patients in the case group experienced a superficial wound infection. The hematoma and cellulitis resolved with nonoperative therapy and oral antibiotics in all cases.
Discussion In this updated, retrospective, case-control study, we found that in addition to warfarin usage, hernia characteristics and
36
j o u r n a l o f s u r g i c a l r e s e a r c h s e p t e m b e r 2 0 1 6 ( 2 0 5 ) 3 3 e3 7
Table 3 e Univariate analysis: cases versus controls. Variable
Odds ratio
95% CI
P value
Valvular disease
11.6
2.6-51.3
<0.01
Incarcerated hernia
11.0
1.3-88
0.01
Bridging with heparin
7.1
2.0-25.1
<0.01
Previous bleed
6.1
1.9-18.4
<0.01
Recurrent hernia
3.7
1.4-9.7
<0.01
Warfarin
3.5
1.6-6.4
<0.01
Atrial fibrillation
2.6
1.2-5.5
0.01
CAD
2.1
1.0-4.4
0.05
Hypertension
2.0
1.1-3.6
0.02
CAD ¼ coronary artery disease.
heparin bridging were independent risk factors for the development of a groin hematoma after IHR. To our knowledge, this is the largest post-IHR hematoma study to date. Whether to continue anticoagulation during elective surgery continues to be a topic for debate. Previous studies suggest that although continuation of warfarin therapy during elective hernia repair does result in higher rates of hematoma, many of these cases do not require operative intervention and can be observed with minimal complications.16 Less than 40% of our hematoma cohort required some kind of evacuation procedure (aspiration or return to the operating room), and less than one-third were found to have a definitive bleeding source in the operating room. In other words, returns to the operating room were not frequent, and when they occurred, a bleeding source was not found in most cases. Based on previous reports and our data, this suggests that continuation of anticoagulation for elective IHR is certainly an option and should be considered for many patients. Another option is to bridge individuals at high risk for thromboembolic complications, during elective operations. This is theoretically done to minimize the risk of complications such as stroke by providing a more transient or more easily reversed type of anticoagulation such as heparin.17 The timing and specifics of bridging have, however, also been debated heavily in the literature.18,19 In a recent multiinstitutional, randomized trial, investigators found that in patients with atrial fibrillation, avoiding bridging with heparin during warfarin interruption for elective operation was noninferior to perioperative bridging.11 In other words, patients who were not bridged were not more likely to suffer from thromboembolic phenomena. In addition, investigators in this trial found that patients, who received bridging, suffered twice the number of major bleeding events. There was heterogeneity in the types of elective operations performed in both arms of the study, but based on these data, avoidance of bridging can safely be considering in some patient groups undergoing elective operations. In our multivariable analysis, incarceration and recurrent hernia were also found to be independent predictors of postoperative hematoma. This may be explained by additional dissection and difficulty in visualization due to scar tissue encountered in these cases. Our data suggest that the type of hernia repair was not a risk factor for hematoma formation.
This is consistent with existing reports that show no major difference in postoperative complications between variations in IHR technique.20-22 These findings highlight the importance of meticulous and thoughtful dissection, particularly in patients undergoing emergent or urgent IHR and those with previous groin operations. Limitations of our study include the single-institution use of retrospectively collected data based on information available only in the electronic medical record. Further analysis of this complication in IHR patients should carefully scrutinize the use of perioperative anticoagulation. A multicenter randomized controlled trial assessing differences in outcome between continued anticoagulation with warfarin, bridging, and anticoagulation cessation is needed for this cohort. In addition, our definition of hematoma was not defined before data collection; this was based on the description provided in the electronic medical record. Finally, our analysis did not take operator (surgeon, surgical trainee) experience and involvement into consideration. It may be that inexperienced surgeons or cases in which surgical trainees had more autonomy, resulted in a higher frequency of hematomas. All procedures involved surgical trainees, but confirming how much of a procedure they performed is not available. Despite these limitations, we believe our data augment the existing literature on postoperative hematoma after IHR. We concur with previous reports that elective IHR in most patients requiring anticoagulation can be performed safely while either holding medications, being bridged with heparin, or in some cases, continuing full anticoagulation. This decision should obviously be tailored toward each specific patient, but the historical approach that all anticoagulation should be ceased before surgery is no longer the norm. Finally, surgeons need to remain vigilant of meticulous hemostasis, especially in patients presenting with incarceration or hernia recurrence. With these tenants in mind, approaching eradication of this rare complication may be feasible in the near future.
Acknowledgment Authors’ contributions: MHZ was involved in study conception, data acquisition, interpretation and analysis of data. TKP and MMEK were involved in study conception and data interpretation. All authors were involved in article drafting and critical revision and approved the final article.
Disclosure The authors declare that they have nothing to disclose.
references
1. Rutkow IM. Demographic and socioeconomic aspects of hernia repair in the United States in 2003. Surg Clin North Am. 2003;83:1045e1051. 2. Sakorafas GH, Halikias I, Nissotakis C, et al. Open tension free repair of inguinal hernias; the Lichtenstein technique. BMC Surg. 2001;1:15.
zeb et al hematoma after inguinal hernia repair
3. Tschudi JF, Wagner M, Klaiber C, et al. Randomized controlled trial of laparoscopic transabdominal preperitoneal hernioplasty vs Shouldice repair. Surg Endosc. 2001;15:1263e1266. 4. Andersson B, Hallen M, Leveau P, Bergenfelz A, Westerdahl J. Laparoscopic extraperitoneal inguinal hernia repair versus open mesh repair: a prospective randomized controlled trial. Surgery. 2003;133:464e472. 5. Kingsnorth AN, Bowley DM, Porter C. A prospective study of 1000 hernias: results of the Plymouth Hernia Service. Ann R Coll Surg Engl. 2003;85:18e22. 6. Fitzgibbons RJ, Giobbie-Hurder A, Gibbs JO, et al. Watchful waiting vs repair of inguinal hernia in minimally symptomatic men: a randomized clinical trial. JAMA. 2006;295:285e292. 7. Smoot RL, Oderich GS, Taner CB, et al. Postoperative hematoma following inguinal herniorrhaphy: patient characteristics leading to increased risk. Hernia. 2008;12:261e265. 8. Sanders DL, Shahid MK, Ahlijah B, Raitt JE, Kingsnorth AN. Inguinal hernia repair in the anticoagulated patient: a retrospective analysis. Hernia. 2008;12:589e592. 9. Mumme T, Peiper C, Lorenz N, Junge K, Schumpelick V. Does anticoagulation increase the risk of wound hematoma in inguinal hernia surgery? Hernia. 2000;4:275e281. 10. Kunkala MR, Kehl J, Zielinski MD. Spontaneous rectus sheath hematomas: when to restart anticoagulation? World J Surg. 2013;37:2555e2559. 11. Douketis JD, Spyropoulos AC, Kaatz S, et al. Perioperative bridging anticoagulation in patients with atrial fibrillation. N Engl J Med. 2015;373:823e833. 12. Mensah PK, Gooding R. Surgery in patients with inherited bleeding disorders. Anaesthesia. 2015;1:112e120. 13. Eker HH, Langeveld HR, Klitsie PJ, et al. Randomized clinical trial of total extraperitoneal inguinal hernioplasty vs
14.
15.
16.
17.
18.
19. 20.
21.
22.
37
Lichtenstein repair: a long-term follow-up study. Arch Surg. 2012;147:256e260. Johansson B, Hallerback B, Glise H, Anesten B, Smedberg S, Roman J. Laparoscopic mesh versus open preperitoneal mesh versus conventional technique for inguinal hernia repair: a randomized multicenter trial (SCUR Hernia Repair Study). Ann Surg. 1999;230:225e231. Choi YY, Kim Z, Hur KY. Learning curve for laparoscopic totally extraperitoneal repair of inguinal hernia. Can J Surg. 2012;55:33e36. McLemore EC, Harold KL, Cha SS, Johnson DJ, Fowl RJ. The safety of open inguinal herniorraphy in patients on chronic warfarin therapy. Am J Surg. 2006;192:860e864. Baron TH, Kamath PS, McBane RD. Management of antithrombotic therapy in patients undergoing invasive procedures. N Engl J Med. 2013;368:2113e2124. Dunn AS, Turpie AG. Perioperative management of patients receiving oral anticoagulants: a systematic review. Arch Intern Med. 2003;163:901e908. Patel JP, Arya R. The current status of bridging anticoagulation. Br J Haematol. 2014;164:619e629. Antoniou SA, Kohler G, Antoniou GA, Muysoms FE, Pointner R, Granderath FA. Meta-analysis of randomized trials comparing nonpenetrating vs mechanical mesh fixation in laparoscopic inguinal hernia repair. Am J Surg. 2016;211:239e249 e232. Koning GG, Wetterslev J, van Laarhoven CJ, Keus F. The totally extraperitoneal method versus Lichtenstein’s technique for inguinal hernia repair: a systematic review with metaanalyses and trial sequential analyses of randomized clinical trials. PLoS One. 2013;8:e52599. McCormack K, Scott NW, Go PM, Ross S, Grant AM. Laparoscopic techniques versus open techniques for inguinal hernia repair. Cochrane Database Syst Rev. 2003:CD001785.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Risk factors for postoperative hematoma after inguinal hernia repair: an update Muhammad H. Zeb, MBBS, T.K. Pandian, MD, Moustafa M. El Khatib, MBBCh, Nimesh D. Naik, MD, Abhishek Chandra, David S. Morris, MD, Rory L. Smoot, MD, and David R. Farley, MD* Division of Subspecialty General Surgery, Mayo Clinic College of Medicine, Rochester, Minnesota
article info
abstract
Article history:
Background: We recently sensed an increase in the frequency of groin hematoma after
Received 12 February 2016
inguinal hernia repair (IHR) at our institution. The aim of this study was to provide a more
Received in revised form
updated assessment of the risk factors inherent to this complication.
19 May 2016
Methods: We performed a case-control study of all adult patients (age 18 y) who developed
Accepted 1 June 2016
a groin hematoma after IHR at our institution between 2003 and 2015. Univariate and
Available online 8 June 2016
multivariable analyses were performed to assess for independent predictors for groin hematoma.
Keywords:
Results: A total of 96 patients (among 6608 IHR) developed a groin hematoma, (60 were
Inguinal hernia
observed, 36 required intervention). The hematoma frequency increased from our previous
Inguinal herniorrhaphy
study (1.4 % versus 0.9%, P < 0.01). Mean age was 64.6 y (range: 18-92), and 84.3% were men.
Postoperative hematoma
There was no significant difference in the laterality, type, or technique of IHR between
Groin hematoma
cases and controls. Univariate analysis (odds ratio [95% confidence interval], P) identified
Anticoagulation
warfarin usage (3.5, [1.6-6.4], P < 0.01), valvular heart disease (11.6, [2.6-51.3], P < 0.01), atrial fibrillation (2.6, [1.2-5.5], P ¼ 0.01), hypertension (2.03, [1.1-3.6], P ¼ 0.02), recurrent hernia (3.7, [1.4-9.7], P < 0.01), and coronary artery disease (2.1, [1.0-4.4 ], P ¼ 0.05) as significant preoperative factors. The proportion of patients on warfarin decreased since our prior report (31% versus 42%, P ¼ 0.20). On multivariable regression, warfarin and recurrent hernia were independent predictors of hematoma development. Conclusions: Independent risk factors for the development of groin hematoma after IHR included warfarin use and recurrent hernia. Careful consideration for anticoagulation and surgical hypervigilance remains prudent in all patients undergoing IHR and especially those with recurrence. ª 2016 Published by Elsevier Inc.
Introduction Inguinal hernia repair (IHR) is one of the most common surgical procedures worldwide. In the United States alone, approximately 800,000 repairs are performed annually.1 Groin hematoma after IHR is an infrequent complication that can
cause significant patient discomfort, require reoperation, and delay postoperative recovery. Several studies reveal that postoperative groin hematomas after IHR occur with a frequency between 0.3% and 6%.2-8 We have previously reported an increase in hematoma occurrence in patients who were on warfarin therapy undergoing IHR.7 A concern has always
* Corresponding author. Department of Surgery, Mayo Clinic, 200 First Street, SW, Rochester, MN 55905. Tel.: þ1 507 284 2095; fax: þ1 507 284 5196. E-mail address: [email protected] (D.R. Farley). 0022-4804/$ e see front matter ª 2016 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jss.2016.06.002
34
j o u r n a l o f s u r g i c a l r e s e a r c h s e p t e m b e r 2 0 1 6 ( 2 0 5 ) 3 3 e3 7
existed regarding the risk of bleeding associated with cessation and resumption of anticoagulation after surgery. Earlier studies have focused on anticoagulation as an important predictor of groin hematoma but have not provided much detail regarding optimal bridging therapy.8-10 Whether bridging with heparin is a contributor to hematoma remains unclear, but a recent randomized study has suggested that bridging triples the risk of bleeding.11 In addition to established coagulation abnormalities, other possible contributors to hematoma formation may include hernia location, recurrence, or incarceration.12 Laparoscopic hernia repair has been proven to have faster recovery and earlier return to work compared with Lichtenstein repair and is the treatment of choice in many patients.13,14 As older surgeons adopt the laparoscopic approach, rates of hematoma may increase as they work upward along the necessary learning curve.15 Herein, we aim to review our experience with groin hematoma after IHR, update our results as previously published, and further analyze risk factors unrelated to anticoagulation.
Methods
with heparin, and anticoagulation restart time frame. In addition, patient medical conditions including presence of cardiac valvular disease, atrial fibrillation, cerebral vascular disease, hypertension, hyperlipidemia, coronary artery disease, pulmonary embolism, hematologic abnormalities (leukemia, polycythemia, and thrombocytopenia) and evidence of previous bleeding were evaluated.
Statistical analysis Univariate analysis was initially used to identify patient and operative characteristic differences between cases and controls. To analyze matched data with a binary outcome, a conditional logistic regression model was used. This model was created so that the proportional hazards for each case and associated matched control were in a separate stratum. The odds ratios (ORs) from these models were reported with 95% confidence intervals (CIs). Multivariable models were then built using stepwise, backward, and best subset selection. All tests were two sided with a significance level of 0.05. All analyses were conducted with SPSS (Statistics for Windows, IBM Corp, Released 2011, version 20.0, Armonk, NY).
Cases and controls
Results Under institutional review board approval, all adults (age 18 y) at our institution who developed a groin hematoma after IHR between the years 2003 and 2015 were studied. These patients were identified using the morbidity and mortality lists for all general surgeons operating at our institution. Morbidity and mortality lists at our institution are generated by each surgical team as complications occur. Despite this “rolling” submission, it is certainly possible that subclinical hematomas not requiring any diagnostic modality or intervention may not have been included in the list. Cases were defined as the patients who developed groin hematoma, confirmed by computed tomography scan or ultrasound, within 30 d of the procedure. To avoid overestimation of hematoma occurrence, patients with clinical findings such as ecchymoses and a mass were not included unless imaging confirmed the diagnosis. Controls were obtained by using surgical indexing to identify all patients who had undergone IHR at our institution during the study years and crossreferencing this with the morbidity and mortality lists to remove those that developed hematoma. Thereafter, each case was matched by age (at the time of repair) and gender in 1:1 fashion. Time of operation for matched cases and controls was within 1 y or less to attempt to match for the period of presentation. This sample size provided approximately 80% power to detect a relative risk of 3.0.
Data extraction Both patient (cases and controls) and procedure characteristics were obtained including age, gender, date of repair, type of hernia (direct, indirect or both), defect laterality, hernia recurrence, incarceration, type of repair performed, date of hematoma formation, concomitant medications (warfarin, heparin, aspirin, clopidogrel), use of perioperative bridging
A total of 96 cases (1.4%) of postoperative hematoma after IHR were identified from 6608 hernia repair operations (Table 1). The mean age for cases and controls was 64.6 y (17.1), and 84.3% were men. There was no significant difference between the cases and controls in terms of hernia laterality (left, right, bilateral; P ¼ 0.85, usage of aspirin [P ¼ 0.15], usage of clopidogrel [P ¼ 1.00], and hyperlipidemia [P ¼ 0.19]). Furthermore, both groups were comparable in regard to hernia repair technique (P ¼ 0.58; Table 2). The mean time to hematoma formation and/or recognition was postoperative day (POD) 4.7 (3.8).
Table 1 e Number of hematomas and hernias performed per year 2003-2015. Year
Hernia repairs
Hematomas
%
2003
566
5
0.8
2004
598
9
1.5
2005
542
10
1.8
2006
506
8
1.5
2007
478
5
0.8
2008
564
6
1.0
2009
588
3
0.5
2010
549
7
1.2
2011
565
7
1.2
2012
472
11
2.3
2013
597
7
1.1
2014
523
12
2.2
2015
174
5
2.8
6608
96
1.4
Total
zeb et al hematoma after inguinal hernia repair
Table 2 e Patient characteristics. Variable
Cases (96)
Controls (96)
64 (18-92)
64 (18-92)
n (%)
n (%)
Male
81 (84)
81 (84)
Female
15 (16)
15 (16)
Average age (range)
Gender
Left
30 (31)
30 (31)
Right
47 (49)
44 (49)
Bilateral
19 (19)
22 (19)
Direct
19 (20)
30 (31)
Indirect
46 (48)
44 (46)
9 (9)
14 (15)
Recurrent hernia
19 (20)
6 (6)
Incarcerated hernia
10 (10)
1 (1)
Type of hernia
Type of repair Bassini
2 (2)
1 (1)
Lichtenstein
57 (59)
58 (60)
Mesh and/or plug
14 (14)
9 (9)
Laparoscopic
23 (24)
28 (29)
Warfarin
30 (31)
11 (12)
Heparin
3 (1)
1 (1)
Aspirin
24 (25)
33 (35)
2 (2)
1 (1)
18 (19)
3 (3)
Valvular disease
19 (19)
2 (2)
Atrial fibrillation
26 (27)
12 (13)
Hypertension
51 (53)
35 (36)
Hyperlipidemia
49 (51)
39 (41)
CVD
14 (15)
4 (4)
Diabetes mellitus
10 (10)
10 (10)
Hematologic abnormalities
26 (27)
14 (14)
Previous bleeding
20 (21)
4 (4)
CHF
10 (10)
5 (5)
Medication
Plavix Bridging with heparin
hypertension (OR 2.0, 95% CI [1.1-3.6]; Table 3). Although not statistically significant, the proportion of patients with warfarin usage decreased since our prior report (31% versus 42%, P ¼ 0.20).7 On multivariable logistic regression, warfarin usage, previous bleeding, incarcerated hernia, bridging with heparin, and recurrent hernia were independent predictors for the development of groin hematoma (P < 0.001).
Anticoagulation
Location of hernia
Both
35
Medical conditions
A total of 41 patients (11 [12%] in the control group and 30 [31%] in the case group) were receiving chronic anticoagulation with warfarin at the time of surgery. Of these 41 patients on warfarin, 21 received perioperative heparin bridging. Only three patients in the control group underwent heparin bridging (with lowemolecular weight heparin). Warfarin therapy was reinstituted on average, on POD 1 (range: 0-4). Preoperative international normalized ratio (INR) values were available for 30 of the 96 cases. Twelve of these 30 cases had an INR value 1.3 preoperatively; the remaining 18 patients had values 1.2. Preoperative INR values were available for 10 of the control patients; seven had an INR 1.3. In our cohort, in elective surgeries, warfarin was stopped 5 days before surgery in 90% of hematoma patients; warfarin was stopped 5 days before surgery in all (100%) control group patients. None of the hematoma patients were on newer anticoagulants (direct thrombin inhibitors such as dabigatran, apixaban, and so forth); however, one patient in the control group was on dabigatran. Due to this small number, these anticoagulants were not included in the univariate or multivariable models.
Cases undergoing evacuation
CHF ¼ congestive heart failure; CVD ¼ cardiovascular disease.
On univariate analysis, history of cardiac valvular disease (P < 0.01), incarcerated hernia (P ¼ 0.01), bridging with heparin (P < 0.01), previous bleeding (P < 0.01), recurrent hernia (P < 0.01), perioperative warfarin requirement (P < 0.01), atrial fibrillation (P ¼ 0.01), coronary artery disease (P ¼ 0.05), and hypertension (P ¼ 0.02) were identified as risk factors for hematoma formation. Cardiac valvular disease had the highest estimated OR (11.6, 95% CI [2.6-51.3]), followed by incarcerated hernia (OR 11.0, 95% CI [1.3-88]), bridging with heparin (OR 7.1, 95% CI [2.0-25.1]), previous bleeding (OR 6.1, 95% CI [1.9-18.4]), recurrent hernia (OR 3.7, 95% CI [1.4-9.7]), warfarin (OR 3.5, 95% CI [1.6-6.4]), atrial fibrillation (OR 2.6, 95% CI [1.2-5.5]), coronary artery disease (OR 2.1, 95% CI [1.0-4.4]), and
Thirty-six patients (38%) had a postoperative hematoma requiring evacuation. Of these, 30 patients underwent reoperation. An active bleeding vessel was identified at the time of evacuation in seven patients. No active bleeding was noted in an additional 20 patients, whereas the remaining three patients were noted to have generalized “oozing” at the time of evacuation. Six patients only underwent needle aspiration of their hematomas.
Morbidity and mortality One patient developed a groin hematoma on POD 6 and was taken to the operating room that day for operative evacuation. Unfortunately, this patient suffered massive blood loss and died on POD 17 due to multiorgan system failure from hemorrhagic shock. Six patients in the case group experienced a superficial wound infection. The hematoma and cellulitis resolved with nonoperative therapy and oral antibiotics in all cases.
Discussion In this updated, retrospective, case-control study, we found that in addition to warfarin usage, hernia characteristics and
36
j o u r n a l o f s u r g i c a l r e s e a r c h s e p t e m b e r 2 0 1 6 ( 2 0 5 ) 3 3 e3 7
Table 3 e Univariate analysis: cases versus controls. Variable
Odds ratio
95% CI
P value
Valvular disease
11.6
2.6-51.3
<0.01
Incarcerated hernia
11.0
1.3-88
0.01
Bridging with heparin
7.1
2.0-25.1
<0.01
Previous bleed
6.1
1.9-18.4
<0.01
Recurrent hernia
3.7
1.4-9.7
<0.01
Warfarin
3.5
1.6-6.4
<0.01
Atrial fibrillation
2.6
1.2-5.5
0.01
CAD
2.1
1.0-4.4
0.05
Hypertension
2.0
1.1-3.6
0.02
CAD ¼ coronary artery disease.
heparin bridging were independent risk factors for the development of a groin hematoma after IHR. To our knowledge, this is the largest post-IHR hematoma study to date. Whether to continue anticoagulation during elective surgery continues to be a topic for debate. Previous studies suggest that although continuation of warfarin therapy during elective hernia repair does result in higher rates of hematoma, many of these cases do not require operative intervention and can be observed with minimal complications.16 Less than 40% of our hematoma cohort required some kind of evacuation procedure (aspiration or return to the operating room), and less than one-third were found to have a definitive bleeding source in the operating room. In other words, returns to the operating room were not frequent, and when they occurred, a bleeding source was not found in most cases. Based on previous reports and our data, this suggests that continuation of anticoagulation for elective IHR is certainly an option and should be considered for many patients. Another option is to bridge individuals at high risk for thromboembolic complications, during elective operations. This is theoretically done to minimize the risk of complications such as stroke by providing a more transient or more easily reversed type of anticoagulation such as heparin.17 The timing and specifics of bridging have, however, also been debated heavily in the literature.18,19 In a recent multiinstitutional, randomized trial, investigators found that in patients with atrial fibrillation, avoiding bridging with heparin during warfarin interruption for elective operation was noninferior to perioperative bridging.11 In other words, patients who were not bridged were not more likely to suffer from thromboembolic phenomena. In addition, investigators in this trial found that patients, who received bridging, suffered twice the number of major bleeding events. There was heterogeneity in the types of elective operations performed in both arms of the study, but based on these data, avoidance of bridging can safely be considering in some patient groups undergoing elective operations. In our multivariable analysis, incarceration and recurrent hernia were also found to be independent predictors of postoperative hematoma. This may be explained by additional dissection and difficulty in visualization due to scar tissue encountered in these cases. Our data suggest that the type of hernia repair was not a risk factor for hematoma formation.
This is consistent with existing reports that show no major difference in postoperative complications between variations in IHR technique.20-22 These findings highlight the importance of meticulous and thoughtful dissection, particularly in patients undergoing emergent or urgent IHR and those with previous groin operations. Limitations of our study include the single-institution use of retrospectively collected data based on information available only in the electronic medical record. Further analysis of this complication in IHR patients should carefully scrutinize the use of perioperative anticoagulation. A multicenter randomized controlled trial assessing differences in outcome between continued anticoagulation with warfarin, bridging, and anticoagulation cessation is needed for this cohort. In addition, our definition of hematoma was not defined before data collection; this was based on the description provided in the electronic medical record. Finally, our analysis did not take operator (surgeon, surgical trainee) experience and involvement into consideration. It may be that inexperienced surgeons or cases in which surgical trainees had more autonomy, resulted in a higher frequency of hematomas. All procedures involved surgical trainees, but confirming how much of a procedure they performed is not available. Despite these limitations, we believe our data augment the existing literature on postoperative hematoma after IHR. We concur with previous reports that elective IHR in most patients requiring anticoagulation can be performed safely while either holding medications, being bridged with heparin, or in some cases, continuing full anticoagulation. This decision should obviously be tailored toward each specific patient, but the historical approach that all anticoagulation should be ceased before surgery is no longer the norm. Finally, surgeons need to remain vigilant of meticulous hemostasis, especially in patients presenting with incarceration or hernia recurrence. With these tenants in mind, approaching eradication of this rare complication may be feasible in the near future.
Acknowledgment Authors’ contributions: MHZ was involved in study conception, data acquisition, interpretation and analysis of data. TKP and MMEK were involved in study conception and data interpretation. All authors were involved in article drafting and critical revision and approved the final article.
Disclosure The authors declare that they have nothing to disclose.
references
1. Rutkow IM. Demographic and socioeconomic aspects of hernia repair in the United States in 2003. Surg Clin North Am. 2003;83:1045e1051. 2. Sakorafas GH, Halikias I, Nissotakis C, et al. Open tension free repair of inguinal hernias; the Lichtenstein technique. BMC Surg. 2001;1:15.
zeb et al hematoma after inguinal hernia repair
3. Tschudi JF, Wagner M, Klaiber C, et al. Randomized controlled trial of laparoscopic transabdominal preperitoneal hernioplasty vs Shouldice repair. Surg Endosc. 2001;15:1263e1266. 4. Andersson B, Hallen M, Leveau P, Bergenfelz A, Westerdahl J. Laparoscopic extraperitoneal inguinal hernia repair versus open mesh repair: a prospective randomized controlled trial. Surgery. 2003;133:464e472. 5. Kingsnorth AN, Bowley DM, Porter C. A prospective study of 1000 hernias: results of the Plymouth Hernia Service. Ann R Coll Surg Engl. 2003;85:18e22. 6. Fitzgibbons RJ, Giobbie-Hurder A, Gibbs JO, et al. Watchful waiting vs repair of inguinal hernia in minimally symptomatic men: a randomized clinical trial. JAMA. 2006;295:285e292. 7. Smoot RL, Oderich GS, Taner CB, et al. Postoperative hematoma following inguinal herniorrhaphy: patient characteristics leading to increased risk. Hernia. 2008;12:261e265. 8. Sanders DL, Shahid MK, Ahlijah B, Raitt JE, Kingsnorth AN. Inguinal hernia repair in the anticoagulated patient: a retrospective analysis. Hernia. 2008;12:589e592. 9. Mumme T, Peiper C, Lorenz N, Junge K, Schumpelick V. Does anticoagulation increase the risk of wound hematoma in inguinal hernia surgery? Hernia. 2000;4:275e281. 10. Kunkala MR, Kehl J, Zielinski MD. Spontaneous rectus sheath hematomas: when to restart anticoagulation? World J Surg. 2013;37:2555e2559. 11. Douketis JD, Spyropoulos AC, Kaatz S, et al. Perioperative bridging anticoagulation in patients with atrial fibrillation. N Engl J Med. 2015;373:823e833. 12. Mensah PK, Gooding R. Surgery in patients with inherited bleeding disorders. Anaesthesia. 2015;1:112e120. 13. Eker HH, Langeveld HR, Klitsie PJ, et al. Randomized clinical trial of total extraperitoneal inguinal hernioplasty vs
14.
15.
16.
17.
18.
19. 20.
21.
22.
37
Lichtenstein repair: a long-term follow-up study. Arch Surg. 2012;147:256e260. Johansson B, Hallerback B, Glise H, Anesten B, Smedberg S, Roman J. Laparoscopic mesh versus open preperitoneal mesh versus conventional technique for inguinal hernia repair: a randomized multicenter trial (SCUR Hernia Repair Study). Ann Surg. 1999;230:225e231. Choi YY, Kim Z, Hur KY. Learning curve for laparoscopic totally extraperitoneal repair of inguinal hernia. Can J Surg. 2012;55:33e36. McLemore EC, Harold KL, Cha SS, Johnson DJ, Fowl RJ. The safety of open inguinal herniorraphy in patients on chronic warfarin therapy. Am J Surg. 2006;192:860e864. Baron TH, Kamath PS, McBane RD. Management of antithrombotic therapy in patients undergoing invasive procedures. N Engl J Med. 2013;368:2113e2124. Dunn AS, Turpie AG. Perioperative management of patients receiving oral anticoagulants: a systematic review. Arch Intern Med. 2003;163:901e908. Patel JP, Arya R. The current status of bridging anticoagulation. Br J Haematol. 2014;164:619e629. Antoniou SA, Kohler G, Antoniou GA, Muysoms FE, Pointner R, Granderath FA. Meta-analysis of randomized trials comparing nonpenetrating vs mechanical mesh fixation in laparoscopic inguinal hernia repair. Am J Surg. 2016;211:239e249 e232. Koning GG, Wetterslev J, van Laarhoven CJ, Keus F. The totally extraperitoneal method versus Lichtenstein’s technique for inguinal hernia repair: a systematic review with metaanalyses and trial sequential analyses of randomized clinical trials. PLoS One. 2013;8:e52599. McCormack K, Scott NW, Go PM, Ross S, Grant AM. Laparoscopic techniques versus open techniques for inguinal hernia repair. Cochrane Database Syst Rev. 2003:CD001785.