| Summary: | 碩士 === 國立臺灣大學 === 臨床醫學研究所 === 91 === Background
Diabetes is a common metabolic disease of multiorgan involvement. The American Diabetes Association reported that the prevalence of diagnosed and undiagnosed DM in the United States is currently 6% and rising. The increase is secondary to a rising incident in obesity, as well as a change in the criteria for the diagnosis of DM from a fasting blood glucose 140mg/dl to 126 mg/dl. Beside impaired blood glucose regulation, many direct and indirect sequelae of DM can occur. Triopathy of diabetes including retinopathy, neuropathy, and nephropathy are familiar to every internist. Lower urinary tract dysfunction is one of the most common complications in this group of patients. However, the entities of diabetic cystopathy are rarely mentioned in the literature.
Disturbances of urinary bladder function in diabetic patients are generally attributed to peripheral autonomic neuropathy. The bladder and proximal portion of urethra are innervated by the hypogastric, pelvic and pudendal nerves. The sympathetic nerves originate at T10-L2. The noradrenergic postganglionic fibers from the hypogastric or pelvic plexus innervate the smooth muscles of the bladder base, internal sphincter, and proximal urethra. The parasympathetic innervation originates in the second to fourth sacral segments and projects to the pelvic plexus. Somatic motor innervation originates in S2-3 and travels to the external urethral sphincter via the pudendal nerve. Chronic hyperglycemia is associated with the loss of myelinated and unmyelinated fibers, wallerian degeneration, and blunted nerve fiber reproduction.
The classic presentations of diabetic cystopathy include impaired bladder sensation, increased cystometric capacity, decreased bladder contractility, impaired uroflow and increased post-void residual. It could be confirmed by cystometrograms, including impaired sensation of first desire to void and a significant increased maximal bladder capacity, a decrease in detrusor contractility. The diabetic cystopathy develops insidiously and symptoms do not appear until the disease is in an advanced stage. The neuropathy of pelvic nerve is believed to be the main factor responsible for diabetic cystopathy. Patients with advanced diabetic cystopathy eventually developed flaccid type neurogenic bladder and are frequently associated with recurrent urinary tract infection.
The precise incidence and prevalence of diabetic cystopathy are difficult to determine because of the insidious onset, discrete symptoms, and differences in the definition of bladder dysfunction. About 43% to 87% of insulin-dependent diabetics developed diabetic cystopathy. Another study showed an average 25% prevalence of diabetic cystopathy in patients on oral hypoglycemic treatment. The correlation between diabetic cystopathy and peripheral neuropathy ranged from 75% to 100%. Nephropathy was seen in 30 % to 40 % of cases.
In clinic practice, it is infeasible to evaluate the bladder function of diabetic patients by complete urodynamic study and nerve biopsy. Hence, we design this prospective study to examine the lower urinary tract function and neurological defects in diabetic patients. Thus, we could find out risk factors of diabetic cystopathy.
Aims, Materials and Methods
There are three aims of this research. The first one is to find out the pattern and physiological meanings of International Prostate Symptom Score in diabetic voiding dysfunction. Thus, we can screen diabetic patients by lower urinary tract symptoms and uroflowmetry. The second aim is to set up the correlation between voiding dysfunction and peripheral neuropathy by current perception test. The third aim is to figure out the risk factors of diabetic cystopathy by logistic regression model. This project was approved by the ethic committee of National Taiwan University Hospital.
During 9 months from September 2002, we enrolled 180 Type II diabetic women in clinic. Male patients were excluded from this study due to the confounder of benign prostatic hyperplasia. Woman with cystocele, history of cerebral vascular accident or who has ever received major pelvic surgery will also be excluded. The patients will be interviewed with a questionnaire comprised of the International Prostate Symptom Score (IPSS) and their clinic histories of diabetes. The IPSS was originally used for quantitative evaluation of subjective lower urinary tract symptoms in patients with benign Prostatic hyperplasia. Subsequently this procedure was extended to evaluate lower urinary tract symptoms in various diseases, including bladder dysfunction, regardless of sex. The symptom index score on the IPSS comprises filling and emptying symptom index scores. Frequency, urgency, and nocturia may reflect the state of filling symptoms, whereas incomplete emptying, intermittency, weak stream, and straining at the beginning of urination may be indicative of emptying symptoms.
In the study, participants would be confirmed no presence of urinary tract infection by urine analysis. All patients in our study were encouraged to intake a lot of water exceeding 500ml. The uroflowmetry and post-void residual measurement were conducted. Patients were requested to void as usual pattern. After the patient voided completely, a catheter was introduced into bladder to measure the residual urine volume, immediately. Patients with a total volume of larger than 500 ml, or residual urine of more than 100 ml, or percent residual volume of greater than 25% are considered to have voiding dysfunction. The urinary symptoms score and uroflowmetry of 244 age-matched healthy females obtained from our data bank were use as control.
Of 180 diabetic women, 57 patients had ever received the current perception test. Among the 57 patients, 15 persons were considered as voiding dysfunction and 42 persons did not. The current perception thresholds were determined for 5Hz, 250Hz, and 2000Hz at the middle finger and great toe in a warm environment. The stimulus was initially increased until a sensation was reported and then short stimuli were applied at progressively lower amplitudes until a minimal threshold for consistent detection was determined. The device has a “dummy“ switch to allow the on/off status of the machine to be concealed from the patient, and forced choice paradigm was used to confirm the minimum threshold for perception. Sensation in the lower extremity was also examined by a light-touch monofilament and a tuning fork.
For statistical evaluation, ANOVA and Bonferroni test were used for lower urinary tract symptoms score and maximal flow rate comparisons. Two-tailed Student’s t test for unpaired data was used for the current perception test. All values in the text and tables give the mean ±standard error of means. In addition, the risk factors predicting diabetic cystopathy were evaluated by multiple logistic regression analysis.
Results
The average age of our diabetic patients is 63 years old. Among the 180 diabetic women, 55（30.5﹪）patients were regarded as voiding dysfunction. One of them developed into decompensation of bladder. Her voided volume was 119ml, but post-void residual volume reached 900ml. Included this one, there were 15 patients thought of high grade of voiding dysfunction due to residual volume more than 150ml. Besides, 3 women were considered as low grade of voiding dysfunction. They have larger bladder capacity without remarkable residual urine. There were 37 patients classified as moderate grade of voiding dysfunction. Of the 55 patients, 18（67.3﹪）had ever received the medical treatment for voiding dysfunction.
Urinary symptoms. These patients with voiding dysfunction had a significantly higher total score (14.2±1.1, mean ±standard error of means ), storage symptoms score (7.4±0.5), and emptying symptom score (6.8±0.7) comparing to the control groups. There were no significant difference in urinary symptoms score between the patients without voiding dysfunction and normal controls. Pearson’s correlation test indicated that the symptom index scores were correlated with residual urine volume (r=0.42, p<0.001). The most common symptom presented in patients with voiding dysfunction was nocturia (71%), followed by weak urinary stream (47.2%). Chi-square test showed the higher proportion of all urinary symptoms in voiding dysfunction group. The IPSS questionnaire included a question on general satisfaction with urinary conditions （QOL index）.When the QOL index score was ≧4, the quality of patient’s life was regarded as poor. Patients with voiding dysfunction had the highest QOL index score （3.8±0.2）.Even the diabetic patients without voiding dysfunction had higher QOL index score （2.7±0.1）than normal controls（1.8±0.1）.
Uroflowmetry findings. Uroflowmetry revealed the significant differences in the maximal flow rate of these groups. Patients with voiding dysfunction had a significantly lowest maximal flow rate（14.5 ±1.3ml/sec）. Diabetic women without voiding dysfunction also have a lower maximal flow rate （19.3±0.7 ml/sec）than normal control group. Patients with an intermittent flow pattern had the higher likelihood of developing voiding dysfunction.
ROC curve. The areas under the ROC curve are presented as a diagnostic performance rate of residual urine volume, lower urinary tract symptoms, and maximal flow rate （Fig.1）. After we chose the adequate cutoff values of these tests, table 5 showed the data for sensitivity, specificity, and the likelihood ratio for the utilization of the three tests. All methods have a high diagnostic performance（area under the ROC curve）.
Current perception test. The results provided good discrimination between the voiding dysfunction and control groups at 5Hz and 250Hz in the foot. Table 6 shows the means of the CPT measurements obtained at all frequency in the finger and toe. There were no significant differences of CPT measurements at all frequency between the voiding dysfunction and control groups in the hand. However, there seems to be a trend showing that the lower frequency of stimuli, the more significant difference between the two groups would be. Sensation test performed by light-touch monofilament and a tuning fork could not show the significance in the two groups.
Risk factors analysis. The multiple logistic regression analysis and stepwise model selection were carried out. The results of risk factors analysis were shown by Table 7. The disease duration more than six years, symptoms duration more than one year, and repeated urinary tract infection in one year were significant in the model selection. Others cannot reach the enough significant difference.
Discussion and Conclusions
The present study evaluated the prevalence of voiding dysfunction in the female patients with diabetic mellitus consecutively sampled in our metabolic department. The prevalence has been reported previously to be 25 to 85%. In most of these reports, the patients were sampled with some confounders, and it was unclear what criteria were used to identify the people with diabetic cystopathy. In our study of 180 consecutive patients, the prevalence of voiding dysfunction was found to be 30.5%, when the residual volume was significantly increased, or bladder capacity larger more than 500ml.This prevalence is similar to that previously reported by Ueda and his colleague（Ueda et al, 1997）.
The presentations of lower urinary tract symptoms in diabetic women with voiding dysfunction differed from those in control groups. Besides, the symptom index was correlated with the residual urine volume. It suggested that the higher lower urinary tact symptoms score disclose the hypocontractility of bladder and post-void residual increased in diabetic women with voiding dysfunction. In addition, the results of uroflowmetry showed the significant differences in the average of maximal flow rate of theses three groups. It means that the development of underactive detrusor in diabetic patients should be earlier more than we think. Even the diabetic patient without voiding dysfunction or symptoms had a lower maximal flow rate than normal controls. The theory about detrusor distention caused decompensation and detrusor failure may be wrong. The areas under the ROC curve of residual urine volume, lower urinary tract symptoms, and maximal flow rate showed their diagnostic performance in diabetic voiding dysfunction. This implies that the symptoms index more than 12 and a maximal flow rate below 15 ml/sec were reasonable criteria for identifying the diabetic patient with voiding dysfunction.
In the results of current perception test, we found the significant difference between the patients with and without voiding dysfunction at lower frequency in the foot. The low frequency detection thresholds correlated with tests of small fiber function, such as C fiber. It suggested that the degeneration of unmyelinated fiber in peroneal nerve would be common in diabetic patient with voiding dysfunction. In the process of lost of bladder sensation in diabetic patients, degeneration of C fiber in pelvic nerve should be important
Risk factors analysis showed the value of history taking in diagnosis of diabetic voiding dysfunction. Diabetic patients suffered from urinary symptoms, repeated urinary tract infection and had a long duration of diabetic history could be clues to indicate the existence of voiding dysfunction. The retinopathy and nephropathy are no correlation with voiding dysfunction in our study.
In conclusion, our data suggested that a high proportion of patients with diabetes had voiding dysfunction of various extents. The earlier occurrence of underactive detrusor plays an important role in the development of diabetic voiding dysfunction. Lower urinary tract symptom score and uroflowmetry evaluations provide a useful tool for the early detection of voiding dysfunction. Peripheral neuropathy detected by current perception test in voiding dysfunction group showed the correlation between neuropathy and voiding dysfunction. Carefully history taking is meaningful in screening of voiding dysfunction in diabetic patients.
|
|---|
