Table 1.

Measurement and methodological issues associated used in the included studies

Assessment method	Measurement issue
Electromyography (EMG)	• Data susceptible to crosstalk, the recording of activity from nearby muscle groups that cannot be distinguished from true pelvic floor muscle activity • Valid comparisons not possible between groups without normalization because of differences in impedance, muscle depth and muscle fiber orientation • Normalization challenging because of possible difficulties in being able to perform voluntary maximum pelvic floor contractions • Normalizing PFM EMG data using MVCs reported to have the smallest standard error of measurement and minimal detectable difference compared to a cough, Valsalva and abdominal crunch maneuvers; however, the participants in the study did not have any form of PFM dysfunction and results may differ in a population with SUI [26] • Artifact from probe movement • Variety of devices used: Periform™ (NEEN Mobilis Healthcare Group, UK) [27–29]; Lifecare PR-02 (Everyway Medical Instruments Co., Taiwan) [30, 31]; VET-A (Nanjing Vishee Medical Technology, Ltd.) [32]; Femiscan™ (Mega Electronics Ltd., Kuopio, Finland) [33, 34]; STIMPON™ (Innocept Biobedded System GmbH) [35–39], a custom probe [25] and unspecified [40] as well as disposable surface electrodes (Mediwatch, UK or Medtronic, Minneapolis, MN) attached to a sponge [41, 42] • While the Periform™ and Femiscan™ have demonstrated poor between-day reliability [43], large coefficients of variation [44] and large standard error of measurement [29], to the authors’ knowledge, the reliability of the STIMPON, Lifecare PR-02 and VET-A probes have not been published in a peer-reviewed journal. The cylindrical tripolar electrode design of the STIMPON would be particularly prone to motion artifact, especially when used during dynamic tasks such as running and jumping [45] • Allegedly good reliability of PFM EMG variables in running across 10 steps in one session based on high ICCs; however, minimal detectable differences were very large (87% of MVC for maximum activity [46]), and high reliability does not reflect validity (i.e., crosstalk contamination), especially as high activity of hip external rotator muscles has been shown while running [47]
Ultrasound imaging (USI) and magnetic resonance imaging (MRI)	• Affected by bladder volume and the multiplanar orientation of the pelvic flor • Affected by posture • Lack of standardized procedures, positions, outcome measures and terminology
Manometry, dynamometry and perineometry	• Embarrassment about leakage might prevent women with SUI straining as directed, especially when assessors not blinded • Difficult to establish whether a true maximum was performed • Intra-abdominal pressure can be misinterpreted as forces having been generated through PFM action [48, 49] • Confounding effect of intra-abdominal pressure greater with perineometry (where air- or fluid-filled chambers inserted into the vagina record pressure changes within the chamber resulting from the sum of the forces acting on all surfaces of the chamber, including forces generated by the descent of the pelvic organs) than with dynamometry [50] • Measures of maximum force-generating capacity may be confounded by poor motor control [33]
Digital palpation	• Subjective and poor reliability [51] • Lacks the sensitivity to gauge small changes in pressure) [52]
Terminology	• Task nomenclature is not standardized • Straining sometimes referred to as during cough or Valsalva maneuver, yet not specified • Tendency for straining and Valsalva to be used interchangeably when they have been shown not to be equivalent [53]: The correct instructions suggested for a Valsalva have been described as: “take a breath, then close the mouth, pinch the nostrils with the thumb and the index finger, then blow air forcefully toward the blocked mouth and nostrils and direct the increasing pressure into the ears” while for straining: “take a breath, then contract the abdominal muscles and strain downwards with the intention to evacuate stool or urine” [53]. Using these instructions, with the Valsalva there was diaphragm and pelvic floor elevation, while with straining there was pelvic floor descent [53]. Another study showed that bladder neck displacement was similar in a cough and Valsalva in incontinent women, but was lower during the cough than Valsalva in both parous and nulliparous controls [54], despite greater abdominal pressures recorded during the cough than during the Valsalva across groups • Furthermore, it has also been shown that the Valsalva maneuver can be accompanied by co-contraction of the levator ani in nulliparous women, affecting measurement of bladder neck descent [20]