A retrospective study on perineal lacerations in vaginal delivery and the individual performance of experienced mifwives

The major finding of this retrospective study is that the midwife is an independent predictive factor for the occurrence of perineal lacerations during a vaginal delivery. The data are somehow in line with the observation of highly varying rates of severe perineal lacerations in midwife-conducted deliveries between different centers [13]. However, concerning severe lacerations, our study did not reveal differences between the individual midwives which might be due to the small sample size. A much larger population would be needed to test a complication with an incidence ranging from 0-2 % per individual midwife.

It could be argued that before addressing quality issues on an individual level, one should compare outcome data on an institutional level. Our rate of overall perineal lacerations (26.2 %) is comparable to published data. A recent multicenter study revealed a rate of 21 % [14]. In an older report an even higher rate of up to two thirds was found [7]. The incidence of severe lacerations of 1.0 % found in our data set is comparable to the results of published data ranging from about 1.0 % [7, 13], to about 4.5 according to the birth center and the risk profile [13, 14]. Moreover, other risk factors for perineal lacerations in our report are in line with the published literature [8, 11, 14]. These include gestational age, parity, and birth weight. It should be emphasized that we evaluated perineal lacerations of all grades, and, thus, a mediolateral episiotomy was protective in our analyses, which is contrary to risk factor analyses for severe lacerations only [8]. In contrast to previous studies that have been reviewed by Pergialiotis et al. [8], the woman’s ethnicity was not predictive which might be caused by the fact that we analyzed all grades of perineal lacerations instead of severe cases only and by the small number of ethnicities other than Caucasian. Moreover, the position of delivery was not of influence in our analysis, which is likely due to the small number of positions other than lithotomy. In order to increase the reliability of our data, we excluded all instrumental deliveries, since not only the midwife, but also the obstetrician, likely exerts maneuvers that could influence perineal integrity in these cases. Thus, we consider it sound to state that the midwife herself was responsible for all manual actions during delivery in the presented cohort study. Moreover, the obstetrician was responsible for inspection of the perineum and, thus, the major outcome parameter should be objective.

All in all, we consider the requirement of benchmarking our results against other published data fulfilled to look at the individual midwife. Since several other parameters were independent risk factors for perineal lacerations in addition to the individual midwife in the multivariate model, we can rule out that some of the midwives cared for more complex deliveries. Thus, differences in handling, probably also in communication with the delivering woman, likely contributed to the differences in outcomes. In our department, midwives usually deliver children based on general recommendations about the last stage of delivery. This includes that the hand of the midwife should control the speed of the head crowning through the vaginal introitus, while the other hand should support the perineum. The mother would then be asked to stop pushing and, when most of the head is out, the perineal ring would be pushed under the neonate’s chin as described previously [12]. It, however, must be noted that literature comparing this “hands-on” method to the “hands-off” or the “hands-poised” methods is controversial. Several studies showed favorable results for the latter two methods [15‐17], others concluded that a clear beneficial effect could be seen for the “hands-on” method [12, 18]. These observations are in line with the last Cochrane database systematic review on this topic stating that “hands-off” versus “hands-on” showed no effect on third- and fourth-degree tears [5] as well as with a recent meta-analysis [19]. Accordingly, current midwifery guidelines recommend that either approach is appropriate [20]. However, randomized controlled trials evaluating different techniques of perineal support may not be the only approach for investigating such complex interventions. Taken these considerations together, it is likely that other modifiable factors contributed to the differences in individual outcomes. Other recommended strategies include emphasis on selective mediolateral episiotomy, good visualization of the perineum at birth and communication with the mother regarding slow pushing [12]. Unfortunately, we could not evaluate the latter two factors due to the retrospective study design. However, in this context, it must be emphasized once more that all analyzed midwives were experienced and had performed several hundred vaginal deliveries before the study period. Moreover, from our experience, the extent to which these maneuvers are exerted differs between the midwives. This is in line with the previously published observations that many midwives preferred a personalized approach rather than an a standardized recommended technique and that many midwives will respond to different clinical scenarios by changing technique [21, 22].

To compare the performance of individuals is not only of interest in order to detect those with a weaker performance and offer them training, but also to identify those with highly favorable outcomes and analyze their method of handling deliveries. Thus, new strategies could be developed that could be tested in subsequent studies. One should learn from the best and this is of a considerably high relevance, especially in such a complex field as a vaginal delivery where many circumstances likely are associated with overall outcome, perineal lacerations among them.

Of course, this study cannot evaluate the midwives’ willingness to revise handling strategies and learn from colleagues. As reported previously for general surgeons, this willingness can be low, and, thus, new strategies might not be adapted, especially among those who consider themselves experts [3]. Whether confrontation of the midwives with their personal rates of perineal lacerations would lead to improvements in outcome shall be the aim of future studies. It is obvious that one would have to handle such an intervention with care and that one individual can receive information only about her own performance with the others’ results remaining blinded.

The retrospective design of the study naturally is a limitation, since several possible influencing factors had not been documented in the electronic system routinely and, thus, could not be evaluated. These include perineal support. However, it was up to the midwife whether to use perineal support or not. The study was based on the assumption of variations in handling strategy on an individual level. Hence, the lack of exact information on perineal support does not decrease the relevance of our major finding, namely that the midwife was an independent predictive factor. One could also be concerned whether perineal lacerations were diagnosed correctly. It has been claimed that rates of perineal lacerations were associated with clinical ascertainment [23] and that the level of recognition could vary [24, 25]. Notably, in a recent Australian report, only 71 % of the midwives reported that they had received training in diagnosing severe perineal lacerations and only 16 % of these felt highly confident in this diagnosis [22]. However, as mentioned in the Methods section, both the midwife and an obstetrician usually assess the integrity of the perineum together for all women which should minimize diagnostic bias. Moreover, one might discuss the clinical relevance of low-grade perineal lacerations. However, these can also cause pain and can even lead to wound-healing problems in rare cases. Nevertheless, our data suggest that some cases of lacerations might be avoided by more careful handling, and, without a doubt, for the mother, the optimal delivery would, of course, be one without any complications at all. We, thus, consider the topic of perineal lacerations of relevance, regardless of the severity.