Hypothesis / aims of study
Multiple system atrophy (MSA) is clinically characterized by a combination of autonomic dysfunction, cerebellar dysfunction, and parkinsonism. Underactive bladder is common in MSA, and the presence of large post-void residuals (PVR) is a core clinical feature in the Movement Disorder Society (MDS)-MSA criteria [1].
The bladder contractility index (BCI), defined as Pdet at Qmax + k·Qmax (k = 5 for males, k = 1 for females), is widely used in clinical practice. However, our previous study demonstrated that Pdet at Qmax + Qmax in female patients did not show a strong correlation with voided percentage (voided volume divided by total bladder volume) [2].
Detrusor contractility can also be evaluated using the Watts factor (WF), defined by detrusor pressure (pdet), detrusor shortening velocity (vdet), and constants a and b as [(pdet + a)(vdet + b) − ab] / 2π. However, WF calculation is difficult to compute in clinical practice. In general, WF can be expressed as A·vdet + B·pdet + C·pdet·vdet + D (A, B, C, and D are coefficients).It may be feasible to approximate detrusor contractility with respect to voided percentage using a regression model based on the structure of the WF equation.
Study design, materials and methods
We retrospectively reviewed 161 patients with MSA (male: n = 88; female: n = 73) who underwent urodynamic studies including pressure–flow analysis.
Because voided percentage ranges between 0 and 1, we assumed that the logit-transformed voided percentage could be approximated by a WF-inspired equation (WF-based approximated contractility):
logit(voided percentage) = A·vdet + B·pdet + C·(pdet·vdet) + D
where vdet was calculated as Qmax / (total volume^(2/3)), and Pdet at Qmax was used as pdet. Multivariable linear regression analysis was performed to determine the optimal coefficients. Pearson correlation coefficients were calculated between voided percentage and BCI, and between voided percentage and the WF-based contractility approximation.
Results
In male patients, the following equation was obtained:
logit(voided percentage) = −2.142005 + 0.004291·pdet + 5.361758·vdet + 0.191002·(pdet·vdet),
where vdet = Qmax / (total volume^(2/3)) and pdet = PdetQmax.
In female patients, the equation was:
logit(voided percentage) = −1.491055 − 0.006559·pdet + 6.214230·vdet + 0.140522·(pdet·vdet).
In male patients, the correlation coefficients between voided percentage and the WF-based contractility approximation and between voided percentage and BCI were 0.705 (95% CI 0.582–0.797) (Figure A) and 0.561 (95% CI 0.398–0.690) (Figure B), respectively.
In female patients, the corresponding correlation coefficients were 0.842 (95% CI 0.759–0.898) for the WF-based approximation (Figure C) and 0.445 (95% CI 0.240–0.613) for BCI (Figure D).
The differences in correlation coefficients between the WF-based approximation and BCI were statistically significant in both sexes (Meng test, p < 0.001).
Interpretation of results
The present findings demonstrate that the WF-based contractility approximation showed a stronger correlation with voided percentage than BCI in both sexes. The improved association may be attributable to two key modifications: (1) replacing Qmax with vdet incoporating bladder volume, and (2) inclusion of the interaction term pdet·vdet reflecting detrusor mechanical power rather than pressure alone.
Because detrusor contractility may depend on bladder volume through muscle stretchs, the use of vdet expressed as Qmax / (total volume^(2/3)) appears physiologically reasonable. Furthermore, inclusion of the pdet·vdet term may better represent the dynamic power-generating capacity of the detrusor muscle.
In advanced MSA, detrusor underactivity predominates, and voided percentage may largely reflect contractile performance rather than urethral resistance. Therefore, a formula incorporating a power-related component may better approximate detrusor contractility than pressure-based indices alone.