Correspondence www.thelancet.com/oncology Vol 12 December 2011 1183 Electromotive administration of mitomycin In the September, 2011, issue of The Lancet Oncology, Savino Di Stasi and colleagues 1 report the feasibility and safety of intravesical electromotive drug administration (EMDA) of mitomycin in non-muscle-invasive bladder cancer before transurethral resection of bladder tumours (TURBT). Their results were excellent; they showed that intravesical EDMA before TURBT reduces recurrence rates and enhances the disease-free interval compared with intravesical passive diffusion of mitomycin after TURBT and TURBT alone. Intravesical BCG is more effective than is intravesical mitomycin for reduction of recurrence rates. In a meta-analysis by Malmström and colleagues, 2 the risk of recurrence with BCG maintenance was 32% lower than with intravesical mitomycin. Urdaneta and coworkers 3 recommend: intravesical chemotherapy for low-risk superficial bladder carci- noma; treatment of intermediate-risk disease with either chemotherapy or immunotherapy with BCG; and treatment with BCG for high-risk tumours. Therefore, the issue is whether intravesical EDMA of mitomycin makes any difference compared with BCG maintenance therapy alone. Di Stasi and investigators 1 have shown that intravesical EDMA of mitomycin is better than intravesical mitomycin alone. Now, the challenge for researchers is to show that intravesical EDMA of mitomycin is better than BCG. Furthermore, a high risk of systemic absorption of the drug exists after TURBT because of the exposed tissue and capillaries. Postoperative intravesical mitomycin instillation is contraindicated by haematuria for the same reasons. Additionally, inadvertent perforation during the subsequent TURBT after intravesical EDMA instillation could lead to chemical peritonitis. Although Di Stasi and colleagues showed that the complication rates of TURBT alone, intravesical passive diffusion of mitomycin after TURBT, and intravesical EDMA of mitomycin do not differ, 1 these concerns still exist. Di Stasi and colleagues should be congratulated for their excellent research into alternative treatment for non-muscle-invasive bladder cancer; I hope that other researchers will continue their good work. I declare that I have no conflicts of interest Christopher C K Ho chrisckho2002@yahoo.com Department of Surgery, Universiti Kebangsaan Malaysia Medical Centre, Jalan Yaacob Latif, Bandar Tun Razak, 56000 Cheras, Kuala Lumpur, Malaysia 1 Di Stasi SM, Valenti M, Verri C, et al. Electromotive instillation of mitomycin immediately before transurethral resection for patients with primary urothelial non-muscle invasive bladder cancer: a randomised controlled trial. Lancet Oncol 2011; 12: 871–79. 2 Malmström PU, Sylvester RJ, Crawford DE, et al. An individual patient data meta-analysis of the long-term outcome of randomised studies comparing intravesical mitomycin C versus bacillus Calmette-Guerin for non-muscle- invasive bladder cancer. Eur Urol 2009; 56: 247–56. 3 Urdaneta G, Solsona E, Palou J. Intravesical chemotherapy and BCG for the treatment of bladder cancer: evidence and opinion. Eur Urol Suppl 2008; 7: 542–47. Pelvic inflammatory disease and ovarian cancer We read with interest the Article by Hui-Wen Lin and colleagues 1 about the relation between pelvic inflammatory disease (PID) and ovarian cancer. The study, which sampled a Taiwanese nationwide population-based database, showed that women with PID, especially those 35 years and younger, were more likely to have developed ovarian cancer than the control population during 1–3 years of follow up. PID, an infection of the upper genital tract, occurs in only 1–2% of women of reproductive age nowadays, 2 which is a substantial decrease in incidence since the 1980s. About 350 000 of the 1 million beneficiaries of the Taiwanese National Health Insurance programme sampled were women aged between 13 and 65 years. 3 Surprisingly, 68 668 women (20%) were identified as having PID, which seems to be a substantial over-representation of PID prevalence. This result was most likely to have been caused by loose inclusion criteria, for two reasons. First, Lin and colleagues classified patients as having PID with International Classification of Diseases, ninth revision, clinical modification (ICD-9-CM) codes 614, 615, and 616. ICD-9-CM codes 614 and 615 refer to inflammatory diseases of the ovary, fallopian tube, pelvic cellular tissue, peritoneum, and uterus (but not the cervix), which are characteristic of PID. However, ICD-9-CM code 616 refers to inflammatory diseases of the cervix, vagina, and vulva, which affect only the lower genital tract and are unrelated or have not yet progressed to PID. Because Taiwan is a tropical country with a high incidence and recurrence rate of vulvovaginitis, inclusion of patients with code 616 diseases is inappropriate. Second, PID usually occurs in young, sexually active women, 2 and rarely occurs in women past reproductive age, who should, therefore, have been excluded from this study. Re- examination of whether patients with PID that is defined by more strict criteria (ICD-9-CM codes 614 and 615, and aged 15–44 years) were still likely to have ovarian cancer and other comorbid medical disorders would be interesting. Oral contraceptives have long been known to decrease the risk of ovarian cancer (relative risk decreases by 20% for every 5 years of oral contraceptive use) by reducing ovulation-induced microtrauma to the ovarian surface epithelium, and other possible mecha- nisms. 4 Women who have used oral contraceptives also have some protection against PID. 5 However, Lin and colleagues reported that the risk of ovarian cancer was raised in patients with PID with a history of oral contraceptive use, although this finding was not significant. John Bavosi/Science Photo Library