Figure 1.
Flow diagram summarizing the search process and results of each phase of the systematic review.
Table 1.
Baseline Characteristics of randomized TEP-and Lichtenstein of all included trials.
Figure 2.
risk of bias summary of all included trials, the eight criteria on the X-axis. Name of first author and year of trial on Y-axis.
+ = adequate. − = inadequate. Blanc = unclear.
Figure 3.
Hierarchy of outcomes according to importance to patients undergoing inguinal hernia repair (GRADE 2004). Some outcome measures may be correlated (e.g. recurrence is included in severe adverse events).
Figure 4.
The Manhattan Overview for benefit and harm.
4a: trials and their outcomes with benefit according to the three dimensions; standard error (SE), graded from patients perspective (0–9) and level of evidence (1a–5). See legends for references to trial numbers I–XIII. 4b: trials and their outcomes with harm according to the three dimensions; standard error (SE), graded from patients perspective (0–9) and level of evidence (1a–5). See legends for references to trial numbers I–XIII. Legend for reading Figure 4 The Roman numbers are corresponding to the clinical trials as stated below. I = Wright 1996 [41]. II = Merello 1997 [25]. III = Heikkinen 1998 [26]. IV = Moreno 1999 [36]. V = Andersson 2003 [28]. VI = Colak 2003 [29]. VII = Gokalp 2003 [34]. VIII = Hildebrandt 2003 [35]. IX = Lal 2003 [38]. X = Neumayer 2004 [37]. XI = Eklund 2006 [30]. XII = Lau 2006 [40]. XIII = Langeveld 2010 [39].
Table 2.
ordering of the available evidence.
Figure 5.
5a: forest plot on chronic pain. Fixed-effect model. 5b: forest plot on chronic pain. Random-effects model.
Figure 6.
Trial sequential analysis of the effect of TEP vs. Lichtenstein anticipating a realistic relative risk decrease of chronic pain of 20% with TEP compared to Lichtenstein assuming a control event proportion of 17% and a type 1 error risk of 5% and a type 2 error risk of 20% (power = 80%). Even in a traditional random-effects meta-analysis the intervention effect is not statistically significant and therefore the cumulative z-curve does not cross the TSMB for harm, constructed for a diversity-adjusted required information size of 14.666 participants either suggesting lack of evidence for TEP reducing the proportion of patients with recurrence.
Figure 7.
7a: forest plot on recurrence. Fixed-effect model. Figure 7b: forest plot on recurrence. Random-effects model.
Figure 8.
TSA on recurrences TEP versus Lichtenstein.
TSA of the effect of TEP vs. Lichtenstein anticipating a realistic relative risk increase of recurrence of 20% with TEP compared to Lichtenstein assuming a control event proportion of 3%, a type 1 error risk of 5%, and a type 2 error risk of 20% (power = 80%). Even in a traditional random-effects meta-analysis the intervention effect is not statistically significant and therefore the cumulative z-curve does not cross the TSMB for harm. The required information size is incalculable due to too little information available, suggesting lack of evidence for TEP reducing the proportion of patients with recurrence.
Figure 9.
TSA on severe adverse events TEP vs. Lichtenstein.
TSA of the effect of TEP vs. Lichtenstein anticipating a realistic relative risk reduction of severe adverse event of 20% with TEP compared to Lichtenstein and assuming a control event proportion of 20% and a type 1 error risk of 5% and a type 2 error risk of 20% (power = 80%). Even in a traditional random-effects meta-analysis the intervention effect is not statistically significant and therefore the cumulative z-curve does not cross the TSMB constructed for a diversity-adjusted required information size of 11.588 participants suggests lack of firm evidence that TEP reduces the proportion of patients with severe adverse events when the analysis adjusts the significance level for considering sparse data and repetitive testing on accumulating data.
Figure 10.
TSA shows more conversions for TEP compared to Lichtenstein.
Table 3.
Checklist of recommendations for future randomized clinical trials, comparing the TEP with the Lichtenstein technique.