Both open and laparoscopic approaches appear to be safe for ventral hernia repair, with low risk for reoperation and low mortality. This finding is particularly valid for open hernia repair since most are performed via an open approach in Sweden.
The considerable difference in numbers between open and laparoscopic procedures performed limited our ability to display superiority of either method. It is important to realise that there is a difference between data generated from prospective trials and those derived from a national register that represents procedures performed in every day practice at a national level.
The risk of death dramatically increases in the group over 80 years after both primary and incisional hernia repair. This calls for precautionary measures, even in technically simple cases.
Numerous variables need to be considered when evaluating the indication for ventral hernia repair. A “watchful waiting” strategy avoids an unnecessary procedure, but has the disadvantage that the hernia aperture might enlarge while waiting, making repair more complicated as well as increasing loss of domain [15]. Furthermore, acute surgery is associated with higher morbidity and mortality.
The outcome of ventral hernia repair depends to some extent on gender [16, 17]. There was a higher risk for reoperation during index admission in women with a primary hernia. The reason for this is unclear and needs further analysis in a prospective setting with hernia-specific variables. Missed hernia, probably due to multiple defects in women with diastasis recti, has been described as a reason for reoperation [18]. Women are also over-represented in injury claims after ventral hernia repair [18]. Previously published papers claim that ventral hernia repair in women has less favourable outcome regarding surgical site infection [19], readmission [20], and chronic pain [21]. The impact of gender in ventral hernia repair is clearly a complex matter. Differences in fat distribution between men and women, a higher proportion of incisional hernias among women (due to caesarean and gynaecological procedures) are possible contributing factors [17]. Furthermore, alterations in hormonal balance during pregnancy affects collagen composition. The midline of the abdominal wall weakens as the components of the extracellular matrix change due to alteration in the progesterone-oestradiol balance and release of corticosteroids. In addition, increased concentrations of the hormone relaxin directly stimulate metalloproteinases that degrade the matrix [22].
Laparoscopically operated patients had a lower risk for acute reoperation during index admission compared to patients operated with an open technique. The lower rate of reoperation could indicate that minimally invasive techniques have lower complication rates. However, bias due to case-mix may play a role here since complicated cases are not always considered eligible for laparoscopy.
Formal conclusions regarding differences between open and laparoscopic hernia repair are difficult to draw from the dataset in the present study. But since there was a higher proportion of incisional hernias in the laparoscopy group and age groups were equally distributed, there was no obvious pattern in the analyses that could explain the lower complication rate in the laparoscopy group in terms of less complex cases or younger healthier patients.
Previously reported data from Swedish and Finnish national patient insurance companies suggest that the risk for inadvertent enterotomy is higher in laparoscopically operated patients [18, 23]. The data presented here, from a much larger cohort, do not seem to support this.
The risk for reoperation and death was somewhat higher after incisional than after primary hernia repair for both open and laparoscopic repairs. This finding is in in line with previously presented data [5, 24]. The fact that an incisional hernia often has a wider aperture than a primary hernia is an important factor contributing to poorer outcome after incisional hernia repair [24].
Patient frailty is a factor to consider when planning surgery [25, 26]. Preoperative optimisation, evaluation of indication, and method chosen for repair are variables influencing morbidity and mortality [27]. The proportion of incisional hernia patients that are frail is higher than among those with a primary hernia.
Nevertheless, we found that age is also a risk factor for mortality after repair of primary hernia, a procedure that is usually technically uncomplicated and the aperture modest.
The findings of the present study indicate that age is the most important risk factor for death. Clinicians should take note and perhaps introduce more stringent indication criteria for repair in different age groups, and that a “watchful waiting” strategy should be more widely applied in the elderly patient group [28].
Mortality and major adverse event rates in the present material could be considered surprisingly low. The fact that patients who had concomitant bowel surgery were excluded from the analysis probably contributed to this. Concomitant bowel surgery during ventral hernia repair probably increases the risk for reoperation, sepsis, and infection [29, 30]. There are also reports indicating that the risk for anastomotic leakage in bowel surgery can be increased by concomitant incisional hernia repair [31]. This is an issue that needs further research to evaluate and understand the underlying pathophysiology involved. Large hernia aperture is an important risk factor for serious complications in ventral hernia repair [15, 24]. Hernia aperture size is routinely registered in specialised quality registers but not in larger registers with a wider focus as used in this study. Though data on size were lacking, most hernias described in this material were probably small.
Analyses of surgical procedures using register data should be regarded with some degree of scepsis [32]. One presumption is that not all patients undergoing ventral hernia repair are eligible for laparoscopic surgery, and that methods are not interchangeable. Our ability to draw conclusions from register data was limited since we could only analyse the variables that currently exist in the register. Comorbidities, choice of mesh, method of fixation, and other variables of interest in hernia research were not available in our database. Randomised controlled trials reflect a controlled situation over a limited time, while epidemiology data reflect routine clinical practice on a population basis. These two sources of information complement one another.
The low number of surgical complications registered may be a sign of high quality, but it may also be due to poor reporting. Reoperation and death are solid variables in Swedish national registers, which is why these two variables were chosen as key outcomes [33]. The overall low mortality and reoperation rates are interesting per se but difficult to compare with existing randomised trials and other published register data. Although data on hernia recurrence were not available for analysis, we did have the number of patients undergoing reoperation after hernia repair during the study period. The 11% reoperation rate in the present study is in line with previously reported data and can thus be an indication of external validity [34].