Evaluation of a Progressive Treatment Program for Erectile Dysfunction in Patients with Diabetes Mellitus

S. Israilov; J. Shmuely; E. Niv; D. Engelstein; P. Livne; J. Boniel


Int J Impot Res. 2005;17(5):431-436. 

In This Article

Patients and Methods

The study group consisted of 284 consecutive patients with DM referred to our tertiary center for the treatment of ED between 1998 and 2002. All patients had DM type II that was diagnosed before the entrance into the study according to fasting glucose level or oral glucose tolerance test. The demographic characteristics of the patients as well as the data regarding the duration of DM and their comorbidities are presented in Table 1 . In all, 43 patients (15.1%) were treated by insulin alone or with combination with oral hypoglycemic agents at the time of the study. The rest of the patients received oral hypoglycemic agents only. DM was relatively controlled.

The diagnosis of ED was based on the International Index of Erectile Function (IIEF).[17] Medical, social and psychosexual history of the patients and their partners were taken. Physical examination, nocturnal penile tumescence test with Rigiscan, penile color Doppler ultrasound and penile electrothermometry[18] were performed. Complete blood count, electrolytes, glucose level, liver enzymes, urea and creatinine, level of prostate-specific antigen, prolactin and total testosterone were checked. No untreated hormonal disbalances in the hypothalamic-hypopituitary-testicular axis were found in our study group, because our department is a tertiary center and all the possible hormonal problems were identified and treated before the reference.

A positive response at all phases of the program was defined as an erection sufficient for vaginal penetration.

Phase 1. Patients without contraindications[19] were offered sildenafil citrate (Viagra), one tablet at 1 h before coitus, 2-3 h after a meal, at a starting dose of 25-50 mg, depending on age. The dose was increased to 50 and 100 mg at intervals of 2-4 days, depending on response, age, and general state of health.

Phase 2. Patients with contraindications to sildenafil, adverse drug effects, or a negative response (to at least 3-5 doses of 100 mg) were offered vacuum therapy with the VED. Details of its applications were explained to the patients individually before commencing. Silicone cream or lubricating gel was applied around the root of the penis, and the cylinder (with constriction rings) was opened and placed on the penis. Negative pressure was applied slowly and gradually over 2-3 min. When the patient reported pain, we removed the cylinder, but left the constriction rings in place for an additional 1-2 min. Two sessions were conducted by the same operators at intervals of 2-3 days. In patients with a positive response, the device was recommended for use at home before coitus.

Phase 3. VED failures were switched to ICI with cocktail of trimix of vasoactive drugs: papaverine 12-25 mg+phentolamine 1.0-2.0 mg+prostaglandin E1 6-25 µg. In patients who complained of pain during erection, the prostaglandin E1 dose was decreased and the doses of the other drugs were increased.

Phase 4. Patients who failed to respond to the ICI were given sildenafil citrate 50-100 mg followed 40-55 min later by a ICI of papaverine 25 mg+phentolamine 2 mg+prostaglandin E1 6-25 µg.

Phase 5. In patients who failed phase 4, we attempted ICI with papaverine, 16 mg, phentolamine 1.5 mg, and prostaglandin E1 15-25 µg, followed 5-10 min later by the VED.

Phase 6. In patients who did not respond to phase 5, we recommended penile implant surgery.

All patients were followed for 2 y. ED treatment was changed during follow-up, as necessary, according to the same 6-phase program. Follow-up consisted of physical examinations for general state of health, IIEF score, and laboratory tests. In patients receiving the ICI, the physical examination also included a search for palpable fibrotic nodules. Penile Doppler color ultrasound and nocturnal penile tumescence test were repeated.

The results immediately after treatment and at the end of follow-up were analyzed according to Bland.[20] Arithmetic mean values and errors of the mean or median values were calculated. Statistical differences were determined with Student's t-test. Values of P<0.05 were considered significant.