NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Treadwell J, Tipton K, Oyesanmi O, et al. Surgical Options for Inguinal Hernia: Comparative Effectiveness Review [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2012 Aug. (Comparative Effectiveness Reviews, No. 70.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of Surgical Options for Inguinal Hernia: Comparative Effectiveness Review

Surgical Options for Inguinal Hernia: Comparative Effectiveness Review [Internet].

Show details


Summary of Key Findings

A comprehensive list of our conclusions appears in Table 18 below. Clinically significant differences are boldfaced (see further discussion below). Data on other comparisons or other postsurgical outcomes were either insufficient to permit conclusions or were not considered major comparisons/outcomes.

Table 18. Conclusions of this review.

Table 18

Conclusions of this review.

For Key Question 1, the only conclusion permitted by the evidence was that QOL, which was reported as “overall change in health status in previous 12 months” at 1 year, was greater after repair than after WW. The evidence on long-term pain interfering with activities was inconclusive due to low precision of the estimate of the difference between groups (OR = 0.42, 95% CI, 0.17 to 1.04, which meant that the effect could be as much as OR = 0.17 in favor of the repair group or it could be 1.04 in favor of the WW group). The actual rates of long-term pain interfering with activities were 2.2 percent in the repair group and 5.1 percent in the WW group. The primary risk of WW is hernia strangulation, but the evidence was also inconclusive on this outcome. The primary benefit of WW is that one avoids (at least temporarily) the risks of surgery.

For Key Question 2, most outcomes favored laparoscopy over open repair, with the key exception of recurrence in the repair of primary hernia, which found slightly lower rates after open surgery (an estimated 2.5 percent for open surgery versus an estimated 3.6 percent for laparoscopic surgery). We considered this to be smaller than a clinically significant difference; however, some patients and clinicians may consider this an important difference. Another way to describe the difference, which may lead one to believe it is an important difference, is in relative terms: an estimated 43 percent higher risk after laparoscopic mesh repair than after open mesh repair in the context of primary hernia. The infrequency of the outcome is why the relative effect sounds larger than an absolute effect.

Most outcomes favoring laparoscopy involved short-term recovery (hospital stay, RTDA, RTW) and certain types of adverse events (lower rates of wound infection and hematoma). Several of these outcomes showed clinically significant differences. These findings are consistent with the typical reasons why laparoscopy is performed. Interestingly, for recurrent hernia, the risk of a second recurrence was actually lower with open surgery than with laparoscopy. Another interesting finding was that long-term pain was less likely after laparoscopy (also a clinically significant difference), which we found for both primary hernia and recurrent hernia.

For Key Question 3, comparing open mesh procedures, many believe that the Lichtenstein procedure is the best option, but we found that for most comparisons, its results are similar to other prominent open procedures such as mesh plug, PHS, open preperitoneal mesh, and the Kugel procedure. The two exceptions to this were when Lichtenstein had better outcomes than mesh plug regarding RTW (a clinically significant difference) and rates of seroma.

For Key Question 4, comparing laparoscopic mesh procedures, the most commonly performed comparison was between TAPP and TEP, and the evidence permitted four conclusions: that RTW is shorter after TAPP and that short-term, intermediate-term, and long-term pain rates are approximately equivalent.

For Key Question 5, comparing types of meshes, we found approximate equivalence for several outcomes of several comparisons. Standard PP mesh had similar rates of recurrence as combination materials. Three types of mesh (standard PP, low-weight PP, and porcine) had approximately equivalent rates of long-term pain.

For Key Question 6, comparing fixation approaches, we found approximate equivalence in recurrence rates for tacks or staples versus no fixation and sutures versus glue. Also, long-term pain was approximately equivalent between sutures and glue and favored the fibrin glue group when compared to staple fixation.

For Key Question 7, many studies have reported that surgical experience lowers the risk of recurrence after laparoscopic repair, but the data were reported inconsistently and do not permit any estimate of the length of the learning curve.

For Key Question 8, no studies have compared surgical exploration to WW, and in a section below we discuss the pertinent issues.

For Key Question 9, comparing laparoscopy to open high ligation, we found that RTDA was similar for both groups. However, the length of stay was shorter after laparoscopy (as one would expect). Also, long-term patient SFN and long-term cosmesis were greater after laparoscopy, and this likely involves the smaller scar.

Implications, Clinical Context, and Applicability

The typical adult in the included studies was a man in his mid-50s, of average weight, suffering from a primary unilateral hernia. About a quarter of the men worked in physically strenuous jobs; for these men, a durable repair is relatively important to prevent a recurrence. Our review can inform numerous treatment decisions these men face. These treatment decisions include:

  • Whether to undergo surgery at an earlier time or wait. Our data was mostly inconclusive on this point; however, we did conclude that QOL 1 year later is better among those who received surgery than those who waited.
  • Whether to choose open surgery or laparoscopic surgery. For primary hernia, we found that some outcomes favor open surgery, and others favor laparoscopy (see summary above). Laparoscopic hernia repair was introduced around 1990.256 DeTurris and colleagues, 2002,257 found that in 1999, the average surgical resident had performed only about 7 laparoscopic hernia repairs during the previous 5 years of residency, whereas the average number of open repairs in the same time frame was more than 50. This suggests that laparoscopic repair did not receive widespread adoption in the first decade after introduction. Furthermore, laparoscopy involves additional considerations including the universal use of general anesthesia and its associated risks, an increase in operation time, and extensive surgical training. These issues should be considered along with the patient outcomes that were the focus of this report.
  • Which type of open surgery. For some outcomes, open procedures yielded approximately similar outcomes, whereas for other outcomes, the evidence favored the Lichtenstein procedure over other open procedures.
  • Which type of laparoscopic surgery. Evidence generally favored TAPP over other laparoscopic procedures.
  • Choosing among meshes or fixation approaches. These surgical aspects are generally chosen by the surgeon based on prior experience and beliefs. Thus, the clinical audience for these questions is surgeons and manufacturers. Another audience would include materials management and purchasing departments. These professionals are often charged with the task of selecting cost-effective products for their facility. The ideal process for selecting mesh types and fixation methods would involve the input of clinicians, materials managers, and purchasing department professionals. We generally found equivalence among a variety of mesh types and fixation approaches. When comparing fibrin glue with staple fixation our results indicate a lower rate of long-term pain with fibrin glue.
  • Consideration of expertise with laparoscopic hernia repair. We found numerous reports that the risk of RC decreases when a more experienced surgeon performs the procedure or when the surgical center has greater procedure volume.

The evidence-based conclusions listed in the previous section are only applicable to the types of patients enrolled in the studies underlying those conclusions. For example, for Key Questions 2 to 7, the large majority of enrolled patients were middle-aged men; therefore, the applicability to women or to older or younger men is unknown. Similarly, for Key Question 9 on pediatric hernia open versus laparoscopic high ligation, both studies excluded patients <3 months old, and so whether the conclusions apply to patients younger than 3 months old is uncertain.

The conclusions we drew for various outcomes, as follows, should be considered within the clinical context:

  • Complications that may occur following surgery to repair an inguinal hernia include recurrence, damage to internal organs (nerves, blood vessels, bladder, and intestines), hematoma, and wound infection. Recurrence can occur up to several years after hernia surgery; this involves a second surgical repair and its corresponding risks of scar tissue, postoperative pain, and organ damage. Risk of wound infection is small and is more likely to occur in older patients or individuals following a more complex hernia procedure.
  • Short-term outcomes such as hospital stay, RTDA, and RTW may be critical for some patients but relatively unimportant for other patients. For example, a man who needs to return to work quickly to support a family may prioritize these recovery outcomes, even at the expense of a somewhat higher risk of RC. By contrast, a retired man may place less emphasis on the short term and more emphasis on a secure repair with minimal chance of long-term pain.
  • Long-term outcomes such as pain and the feeling of a foreign body are typically important for all patients undergoing inguinal hernia mesh repair and fixation. The main goal of mesh repair is to help strengthen the abdominal wall, while the fixation method used should keep the mesh secure. If an avid runner or cyclist has a mesh repair and experiences long-term pain and/or feels a foreign body or substance, exercise may become uncomfortable. Sedentary individuals may be less concerned with these outcomes. Therefore, some factors of importance for individuals undergoing mesh repair include the mesh material, the rigidity or flexibility of the mesh, whether the mesh allows tissue in-growth, and the method of mesh fixation.


One limitation of this review is that we included only studies published in English. Many studies have been published in other languages, and the inclusion of those studies may have resulted in additional conclusions or may have contradicted some conclusions. To address this possibility, we summarized the abstracts from non-English language publications that may have potentially been included for each Key Question. We also provided citations for these articles so that interested readers can obtain these articles and determine the possible impact had they been included.

Even though we required English-language publication, 76 percent of the studies we included were conducted in countries whose primary language is not English. Thus, many researchers probably chose to translate their work into English. It is unclear whether researchers perform English translation for all of their studies; if not, the translated studies may not fully represent the literature.

Another limitation of this review is that the evidence was inconclusive because of low precision for many outcomes. In general, the included studies were well-conducted but small. We maximized the power of the data by conducting meta-analyses wherever appropriate and possible. Nevertheless, the data often precluded conclusions because they suggested contradictory conclusions (i.e., that the evidence could favor option A or B by a clinically significant amount). The problem was insufficient enrollment, not a lack of followup of enrolled patients, because most studies reported data on at least 85 percent of enrolled patients.

A third limitation is that no studies met our inclusion criteria for Key Question 8 on pediatric contralateral hernia. No studies have compared surgical exploration with WW in this population. Therefore, in the next section, we describe informally some of the existing research in this area, such as the percentage of pediatric patients with a unilateral inguinal hernia who have a contralateral patent processus vaginalis (CPPV), which is a risk factor for inguinal hernia.

Pediatric Contralateral Hernias

As noted, our searches included no studies for Key Question 8, which involved pediatric inguinal hernia and whether to surgically explore for a contralateral hernia or use a wait-and-see approach. This section discusses the pertinent clinical issues.

Some pediatric patients with a unilateral hernia that requires surgical high ligation may develop hernia on the contralateral side later in life. Key Question 8 addressed whether same-operation exploration/high ligation differs from WW in health outcomes or adverse events among these patients. Some surgeons suggest performing routine contralateral groin exploration/high ligation during the operation for unilateral hernia.258-260 The potential benefits of the same-operation exploration/high ligation include the elimination of the need for a second operation (as well as a second anesthesia) for high ligation of a contralateral inguinal hernia and minimization of the risk of incarcerated contralateral hernia (as well as its associated morbidity).259,261

Contralateral hernia exploration/high ligation can be achieved via open surgery or laparoscopic approaches (e.g., transinguinal via umbilical route or via hernia sac).258,262 Laparoscopic approaches became increasingly popular in recent years due to their lesser invasiveness, which may help reduce the risk of damage to the spermatic cord structures.259,261 Some clinicians also believe that laparoscopic approaches allow a more accurate evaluation of the presence of CPPV.262

During the contralateral groin exploration, a large percentage (more than 30 to 40 percent) of the patients were found to have patent processus vaginalis,261,263 which is a risk factor for the development of inguinal hernia. This finding is one of the main reasons for some researchers to recommend routine same-operation contralateral hernia exploration/high ligation.258,263 These researchers also argue that the procedure is generally simple, quick, and safe, rarely causing severe complications.258,263

However, not all CPPV identified during the contralateral exploration will develop into a clinical hernia. Despite the high percentage of the CPPV identified, the incidence of contralateral hernia is fairly low, from 5.6 percent to 11.2 percent (followup of 2 to 29 years), according to several studies.264-268 In addition, a small risk of complications is associated with the exploration/high ligation, including damage to vas deferens and spermatic vessels, recurrences of hernia, or iatrogenic cryptorchidism (undescended testes).264,269,270 Because of the low incidence and potential risk of complications, more researchers believe, routine contralateral groin exploration is not justified and unnecessary.258,259,264-268

From the perspective of patients or families, the decision on whether WW or same-operation contralateral exploration should be chosen would always involve trade-offs among the benefits and risks that are potentially associated with the two treatment options. The ideal study design for addressing Key Question 8 is an RCT in which the researchers assign patients with unilateral hernia randomly into a same-operation exploration/high ligation group and a WW group, follow up the patients for a long period of time after the intervention, and then compare health outcomes that reflect the tradeoffs the patients have to make (e.g., QOL and patient or parent SFN). For this evidence review, we also accepted nonrandomized, prospective comparative studies that made appropriate adjustment on key baseline difference between the two treatment groups. However, our literature search did not identify any studies that met the inclusion criteria for Key Question 8. None of the clinical studies that we have scanned concurrently compared same-operation exploration with WW.

While RCTs are desirable for addressing Key Question 8, it could be technically challenging to conduct this type of study because of the extremely long period of followup that would be required and anticipated difficulty in patient recruitment. Studies examining the risk of developing contralateral hernia for patients by age, sex, and side of the symptomatic hernia would be helpful. Data on the incidence of various adverse events associated with either treatment option (e.g., incidence of strangulated or incarcerated hernia among patients on WW and the incidence of surgery-related complications among patients undergoing same-operation exploration) would also be required to assist patients and families make decisions.

Future Research

We identified several gaps in the evidence in the course of conducting this review. To characterize the gaps, we examined the 87 comparisons and outcomes for which the evidence was insufficient to permit a conclusion, and determine what were the primary reasons for the rating of Insufficient. In 31/87 cases (36%), the only component preventing a conclusion was imprecision. Thus, quite often, there were simply not enough studies and/or the studies had insufficient patient enrollment. In a further 51/87 cases (60%) of the cases, there was a problem with consistency as well as precision. Problems with consistency involved either the existence of only a single study (and therefore the inability to assess consistency) or conflicting results among multiple studies. In the remaining 4 cases, precision was sufficient, yet there were problems with both consistency and selective outcome reporting.

A large portion of the existing literature on inguinal hernia has been conducted outside the United States. The differences in health care systems and practice patterns between the United States and other regions might have an impact on the applicability of the evidence from the perspectives of U.S. stakeholders. Future U.S. studies should define the unique needs of the U.S. population, describe how its needs may differ from those of Europeans (who comprise the majority of patients in studies conducted outside the United States), and target research to these unique areas. Surgical registry may help define unique needs, but existing registries may be inadequate because they are voluntary. For example, an analysis of the voluntary Society of American Gastrointestinal and Endoscopic Surgeons database found that it contained only 1,607 inguinal hernia repairs in a 5.4 year period (September 1999 to February 2005).271 This is about 300 per year, which is a very small portion of the annual U.S. repairs (which has been estimated at 770,000).5

A large registry might also help address the widespread problem of imprecision mentioned above. Many randomized trials have investigated important questions, but their modest size limits the usefulness of the data. Rare events such as RC require much larger sample sizes to permit clear inferences. Registry data require sophisticated analytic techniques, such as propensity scores or instrumental variables, to reduce selection bias. The registries that we assessed (e.g., Swedish Hernia Registry) were quite large (e.g., 174,000 hernias), but authors did not utilize these techniques; therefore, it was difficult to determine the potential impact of selection bias.

Another problem with registry data is the difficulty users would have in determining whether the assessment of RC involved a patient visiting a clinic or simply involved self-reporting via a telephone interview or questionnaire. Most of the studies we reviewed (i.e., not of registries) had patients come into the clinic for a physical assessment, rather than rely on patient reports of recurrence.

Another key focus of future research should be on recurrence rates in the very long-term. The typical patient was middle-aged, so he likely has a few decades of life ahead. Studies have not generally reported recurrence rates past 5 to 10 years, but conceivably patients and clinicians would be interested in much longer timeframes (e.g., 25 years). One surgeon272 proposed projection factors to predict 25-year recurrence risk from specific short-term rates: to estimate the 25-year recurrence rates, multiply the 1-year rate by 5; multiply the 2-year rate by 2.5, multiply the 5-year rate by 1.5; and multiply the 10-year rate by 1.2. These projection factors reveal that a small difference in the short term can correspond to a large difference in the long term (e.g., 1 percent versus 2 percent risk at one year corresponds to 5 percent versus 10 percent risk at 25 years). These projections are a step in the right direction, but they have not been tested empirically. We also encourage greater focus on outcomes that matter most to patients, such as chronic pain, long-term QOL, SFN, and the feeling of a foreign body. These outcomes may be associated with the type of mesh or mesh fixation methods, but our evidence review neither revealed nor ruled out key components, due to low precision.

Some outcomes after hernia repair are defined and measured differently by different authors, and the field would advance more rapidly if more standardization were employed. For example, many studies reported the amount of time before returning to work, but rarely did studies describe how these data were collected. Some may have asked patients for their estimate of the time, and others may have clarified whether the return to work was unrestricted or involved some physical limitations (e.g., a person with a mixed manual/office job may have returned to work only for the office portion of the job and only later could perform the manual labor aspects). Similar comments apply to the outcome of RTDA; what was actually measured and how it was measured could be standardized so that different studies could be more easily compared.

Below, we summarize additional future research needs separately for each Key Question.

We identified only two studies that met the inclusion criteria for Key Question 1. Both studies compared mesh-based open surgery with WW. We identified no study comparing laparoscopic repair with WW. In regards to the two studies included for review, the strength of the evidence on the outcomes reported is rated as either Insufficient or Low. One of the studies reported outcomes only up to 12 months, which is barely sufficient for comparing the two interventions. A need exists for future high-quality studies that compare hernia repair—particularly laparoscopic repair—with WW. These studies should place more emphasis on the outcomes not reported in existing literature or that were insufficient to permit conclusions, such as long-term pain limiting daily activities.

For comparing open and laparoscopic hernia repair, future studies would be easier to interpret if surgeons' prior experience with the study procedures were similar. Long-term recurrence rates would be expected to be higher for procedures performed with less prior experience. For primary hernia comparing open repair versus laparoscopy, the evidence was sufficient to permit conclusions for several outcomes, but for bilateral hernia and recurrent hernia there was far less evidence and therefore fewer conclusions. Recurrence data has often been reported at median followup (e.g., x percent recurrence, patients had been followed for a median of 2 years with a range from 1 month to 7 years), but given the wide range of followup, this is more difficult to interpret than recurrence data at specific time points (i.e., x percent recurrence at 2 years after surgery).

Another issue for open and laparoscopic repair concerns the mode of anesthesia (local or general). Laparoscopic repair invariably involves general anesthesia, whereas open mesh repair can involve any type of anesthesia. This difference could potentially explain any short-term differences in postoperative pain, if the anesthesia mode has any lingering effects. Future studies should consider comparing modes of anesthesia to determine its impact.

We identified 21 studies that met the inclusion criteria for Key Question 3. Only one of the studies was conducted in the United States (see discussion above). Meanwhile, given that the strength of the evidence on most of the outcomes reported is rated as Insufficient or Low, future studies should be conducted to address the uncertainty with the evidence on these outcomes, particularly recurrence, long-term pain, and severe adverse events.

Only 1 of the 11 studies included for review for Key Question 4 was conducted in the United States. Again, the differences in health care systems and practice patterns between the United Sates and other regions might have an impact on the applicability of the evidence from the perspectives of U.S. stakeholders. There is a need for more studies conducted in the United States. Meanwhile, the strength of the evidence on most of the outcomes reported in the evidence base is rated as Insufficient or Low; future studies should be conducted to address the uncertainty with the evidence on these outcomes, particularly recurrence, long-term pain, RTDA, RTW, and severe adverse events.

The largest literature base for Key Question 5 was found for the comparisons of PP mesh with combination material mesh (17 studies), and for the comparisons of tacks or staples versus no fixation (7 studies) and sutures versus glue (7 studies) for Key Question 6. One of the issues with the literature base for these Key Questions that prevented all studies from being combined in a meta-analysis for a specific outcome was how the data were reported. Outcomes of interest for these comparisons include recurrence, the feeling of foreign body, infection, and pain assessed long-term; however only a few studies reported these outcomes. The size and severity of the patient's hernia have some bearing on the outcomes of the procedure, such as recurrence. Of the 56 studies identified for Key Question 5 and Key Question 6, only 12 studies reported the hernia size for the included patients. Also, not all studies reported data in a similar format (e.g., number of patients versus number of hernias treated). Meta-analysis was based on the ability to combine the reported data. Theoretically, some meta-analyses might have changed if all data had been reported in a similar format. Future research should work to report data in a consistent manner and continue to assess differences in mesh materials and fixation methods.

Regarding Key Question 7, future studies of the relationship between laparoscopic surgical experience and subsequent RC should control for a possible time confound (see discussion of this point above) by reporting recurrence data at a specific time point (e.g., the 2-year recurrence rate was x percent for hernias repaired in the first half of the series and was y percent for hernias repaired in the second half of the series). Studies that investigate surgical experience by comparing different centers or comparing different surgeons need to ensure that RC is measured in the same way across centers or surgeons and that the case mix is similar (to rule out the possibility that more experienced surgeons had lower recurrence rates because they operated on lower-risk patients).

We did not identify any studies that met the inclusion criteria for Key Question 8, WW versus surgery in pediatric populations. The ideal study design for addressing Key Question 8 is an RCT in which the researchers randomly assign patients with unilateral hernia into a same-operation exploration/high ligation pair group and a WW group, follow up the patients for a long period of time after the intervention, and compare health outcomes that reflect the tradeoffs the patients have to make (e.g., QOL, patient or parent SFN). While RCTs are desirable for addressing Key Question 8, it could be technically challenging to conduct this type of study because of the required extremely long period of followup and anticipated difficulty in patient recruitment. In the near future, non-RCT studies are welcome. These studies should focus on the identification of the subpopulation (by age, sex, and left or right side of the unilateral hernia) at high risk for developing contralateral hernia. Future studies should also further investigate the incidences of severe adverse events associated with either same-operation exploration/high ligation (e.g., surgery-related complications) or WW (e.g., strangulated or incarcerated hernia). The findings of these studies would provide crucial information that patients need for making the choice between the two treatment options.

We identified only two studies that met the inclusion criteria for Key Question 9: does open hernia repair without a mesh differ from laparoscopic hernia repair without a mesh in pediatric patients? Both studies enrolled fewer than 100 patients, and both were conducted outside the United States. Only five health outcomes of interest were reported in the two studies, and the SOE for most of these outcomes is rated as either Insufficient or Low. There is a need for future U.S. studies with larger enrollment that address the uncertainty in the evidence on these outcomes. Studies are also needed to address the outcomes that were not reported in the evidence base (e.g., QOL, patient/parent SFN).


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...