We assessed the included original studies with the JBI, and the results are presented in Table 2. Only one study was given a perfect score [30], and the remaining studies presented various methodological issues limiting the external validity of their reported chronic pain rates. We evaluated the three included systematic reviews using the AMSTAR2, which resulted in a grading of “critically low quality” for all three [23, 27, 31]. Detailed results are available in Table 3.

A major reason for the heterogeneity and inaccuracy in the reported chronic pain rates is that the temporal aspect is disputed. Chronic pain is most commonly defined as pain persisting either three or 6 months postoperatively [9], but a 1-year threshold has also been proposed [8]. The international treatment guidelines have not yet agreed on this [5, 6]. A 3-month threshold is in line with the original IASP definition of chronic pain [36] and the recommendations by the HerniaSurge Group [5], but some argue that 6 months are necessary after mesh-based hernia repairs to allow for the mesh-related inflammatory response to decrease [4]. The included studies using a 3-month threshold reported chronic pain rates ranging from 16%–54%, and the studies using af 6-month threshold reported rates ranging from 10%–23%. In hernia surgery, a 6-month threshold may be a more accurate reflection of the pathophysiological transition from acute to chronic pain, even though the exact mechanism behind this transition is not entirely clear [4–6, 27]. The aetiology of chronic pain after inguinal hernia repair is likely to be multifactorial, but it is mostly thought to be of neuropathic origin, which justifies the extensive attention given to intraoperative nerve management [5].

The ICD-11 definition [40] and the current IASP definition specific for chronic postsurgical pain [41] maintain that chronic pain is pain persisting 3 months postoperatively. This threshold is standardised across all surgical procedures, but seemingly for pragmatic purposes rather than pathophysiological [40], and these definitions do not seem to explicitly consider the potential influence of implants, such as meshes. Most important, however, is that an operational definition for this is determined through scientific consensus.

In 2020, IASP updated its general definition of pain [42]. The new definition maintains an emphasis on pain as a personal subjective experience, but it also slightly downplays the role of tissue damage as the source of pain. However, this is unlikely to have direct implications for the definition of chronic pain, but we are not aware of any reports or responses from the hernia research community on this yet.

In the included studies, there was a large variation in the length of follow-up (total range of 3–84 months). However, it is important to note that postoperative pain declines substantially over time [38], and with a distinct decrease at around 3.5 years postoperatively for laparoscopic repairs [43]. This may account for some of the reported variation, and many of the chronic pain rates reported in the included publications are in fact incomparable because of the temporal decline in pain.

Many different rates of chronic pain have been disseminated throughout the literature and have been reproduced in hundreds of papers, often with reference to the publications included in this review (Table 1). However, most of these rates are based on reports of pain of any severity, i.e., “any pain”, which arguably is a substantial overestimation of what matters to the patients, and that is presumably a level of pain that interferes with their daily activities or their quality of life. These concepts are more complicated to assess accurately, and they require validated multi-dimensional tools. They are likely to be more meaningful and consequential for patients, however, firm evidence of patient preferences on these issues is still lacking, and it is a topic that surely needs further research.

Two of the included clinical trials applied “any pain” as a primary endpoint, but none applied moderate-to-severe pain as primary endpoint, which may be criticised [18, 28]. Even though the rate of moderate-to-severe pain is a rarer event requiring a greater sample of patients to achieve a high enough power, it is likely a more important and meaningful endpoint for the patients [44, 45]. A sufficient sample size can be achieved by multicentre or international collaboration.

The majority of the studies selected for this review [16–20, 22–24, 26, 28, 30, 31, 34, 35] reported exceptionally high rates of chronic pain in their respective abstracts, where most of them were based on the “any pain”-definition. Many of the publications also reported rates of significant/severe pain with lower associated rates, however, the higher rates of “any pain” reported in abstracts tend to be reproduced in the literature. It could be speculated that a fear of publication bias might have influenced this, and this may contribute to an inflated perception of the extent of the problem of chronic pain.

The measurement of pain is by definition indirect and complex [46], and most of the included studies suffer from issues regarding the measurement of pain. A factor contributing to the variation of reported chronic pain rates is the abundance of different methods of measurement available, of which many are suboptimal [47]. Some studies use unidimensional tools, such as a verbal rating scale or a visual analogue scale, which may be excellent for measuring acute pain but are generally not considered sufficient measures for the complex nature of chronic pain [46, 48]. Unvalidated and unidimensional instruments are an unreliable approach, and validated multi-dimensional measures are necessary [46].

Another important aspect is the timing of measurements: when is pain measured and which timespan are patients asked to account for. Assessing a patient’s current level of pain is only a snapshot, which is susceptible to random interference from unrelated factors without reflecting the natural day-to-day fluctuations [46]. Alternatively, patients could be asked to assess pain experienced during a preceding period of a specified length (days/weeks/months), which is probably a more stable measure, but it does entail a risk of recall bias. Retrospective assessment of previously experienced pain is even more at risk of recall bias, as pain memory is notoriously unreliable, and it is dependent on numerous contextual factors including a patient’s current experience of pain [46, 49]. The preferable timing of measurements remains a topic for further discussion, however, prospective measurements should be favoured [50].

A highly speculative aspect of pain measurement is that patients might overexpress their experience of pain as a result of being surveyed about it. This is known as a negative Hawthorne effect and might account for a small portion of the high rates of chronic pain reported in the literature [51].

In this review, we have selected the most cited and impactful publications reporting on the rate of chronic pain. Evidently, most highly cited studies are older, and the most recent study included here was published in 2011 [35]. However, all of the included studies remain frequently cited to this day, which is remarkable considering the rapid developments in the quality of hernia surgery in recent years [12]. A majority of the included studies involved only or predominantly open repairs, but these publications do not reflect current surgical practice in many developed countries, as laparoscopy has become increasingly popular [37] and is associated with a lower risk of chronic pain [38, 52, 53]. As described, several factors are contributing to higher quality in surgical practice, which hopefully have already translated into lower chronic pain rates. With all of this in mind, the conventionally reported rates of chronic pain are likely outdated.

Currently reported chronic pain rates likely overestimate the actual rate in 2022 or at least the rate of pain of a degree that is meaningful to patients. These seemingly excessive rates are widely reported in the literature, they are easily available to patients online [54], and patients are routinely informed about them during preoperative visits. Presumably, this could deter patients from receiving necessary surgery if they are scared off by potentially obsolete and misleading information.

In general, an accurate estimate of the extent of a problem is necessary to determine proportionate preventive measures and decide appropriate therapeutic efforts [7]. This is certainly also true in the case of chronic pain after inguinal hernia surgery. A more accurate and contemporary estimation of the chronic pain rate is critically necessary to better inform shared decision-making between patients and surgeons based on current and high-quality evidence. Ultimately, patients deserve an accurate basis for an informed choice about their own treatment.

In addition to an overall chronic pain rate, it might also be informative and valuable for patients to assess and communicate demographic-specific rates whenever possible. The risk of chronic pain is dependent on multiple different demographic variables, and predictive modelling and individualised risk estimation partially based on these may be a beneficial approach going forward [55].

This study is an informal review of 20 high-impact studies. It is not a comprehensive systematic review of the entire literature, and due to the nature of this study, it does not include more contemporary research. For this, we refer to the literature [38, 39].

The applied search engine, Google Scholar, is an excellent source of freely available bibliometric data, however, it does provide some limitations. In general, Google Scholar identifies more citations than comparable search engines, and it includes more non-journal citations (i.e., books, conference abstracts, grey literature, etc.) [56], which could lead to a risk of overestimating the actual impact of the included studies. Google Scholar utilises an effective but also non-transparent search algorithm, meaning that literature searches have poor reproducibility, which is a limitation of this search engine [56, 57]. Google Scholar only allows for access to the 1,000 most relevant search results for any particular search query, and thus, the literature search in this review cannot be considered entirely comprehensive [58]. However, the search algorithm determines the relevance of search results partially based on citation count, which effectively minimises the risk of missing any highly cited records [59].

In this review, the inclusion of studies was partially based on the overall citation count of each study. This was a pragmatic approach, and the number of citations is only a surrogate measure of study impact or quality, and the reliability and accuracy of this measure is debatable. Inclusion was also based on number of citations per year since publication, which was an ad-hoc measure implemented to ensure contemporary relevance of the included studies and to avoid including no longer cited studies. However, both of these inclusion criteria discriminate against newer publications. This study was limited by the fact that the applied thresholds (>100 overall citations, >10 citations per year) were entirely arbitrary. Furthermore, this review may be vulnerable to selection bias, as the included highly cited studies may have been more prone to some of the methodological pitfalls described above than studies with lower citation counts.

We selected the JBI for the risk of bias assessment in this review, because it is a tailored tool for studies reporting prevalence [14]. However, it should be noted that many such tools exist [60, 61], and it has previously been suggested that the JBI may not be an optimal tool, but it was included in the present study as it is likely the best available tool for this purpose [60, 61]. The JBI is designed exclusively for the evaluation of prevalence data, and the results produced by the JBI should not be interpreted otherwise. The poor evaluation of most of the included studies (Table 2) is primarily a reflection of the fact that most of these studies were likely never intended to produce generalisable estimates of prevalence or incidence. Accordingly, the presented results indicate nothing about other aspects of the studies. An additional limitation of the JBI is that it does not provide an overall score for each evaluated study, which can make it difficult to operationalise its results in a systematic review or meta-analysis.

For the reasons above, we do not claim this study to be exhaustive. Nevertheless, the results and conclusions remain valid for the included studies.

In this review, we have explained and demonstrated that the chronic pain rates conventionally reported after inguinal hernia repair in the literature are obsolete, probably inaccurate, and likely exaggerated. This is due to uncertainties about the definition and measurement of chronic pain, other methodological shortcomings, and the fact that recent advances in inguinal hernia surgery are not reflected in the included publications. We have also highlighted the importance of solving these issues by determining consensus-based definitions and standards, and subsequently performing large, well-designed studies to establish a more accurate chronic pain rate. For this, we need prospective multicentre studies that apply clear evidence- and consensus-based definitions, use validated measurement instruments, and are reflective of current surgical practice and quality.