M agnetic-resonance-guided percutaneous laser ablation of uterine fibroids

reduction was of similar magnitude to that predicted by previous population-based and preliminary intervention studies.2,4 The mechanism for the reduction remains undetermined. High-dose intra-arterial ascorbic acid may ameliorate endothelial dysfuntion in hypertensive patients,5 although we found that ascorbic acid treatment had no effect on systemic markers of nitric oxide bioactivity, or prostacyclin production. There also was no evidence that ascorbic acid reduced lipid peroxidation and production of vasoconstrictor F2 isoprostanes (8-epi-prostaglandin-F 2␣). However, these systemic markers are limited because they may not accurately reflect events in the vascular wall. Although confirmation of these findings in larger, longer studies is required, the present study suggests that 500 mg of ascorbic acid daily is useful for blood pressure control in patients with hypertension. Supported by grants from the National Institutes of Health (HL53398, HL55993, HL59346, HL55854, HL03195, and HL56170). NG is the recipient of a Fellowship Award and JAV is an Established Investigator of the American Heart Association. We thank Liza Hunter, Nirav Shah, and Mariusz Olesiak for technical help. 1

2 3

4

5

Nakazono K, Watanabe N, Matsuno K, Sasaki J, Sato T, Inoue M. Does superoxide underlie the pathogenesis of hypertension? Proc Natl Acad Sci USA 1991; 88: 10045–48. Salonen JT, Salonen R Ihanainen M, et al. Blood pressure, dietary fats, and antioxidants. Am J Clin Nutr 1988; 48: 1226–32. Enstrom JE, Kanim LE, Klein MA. Vitamin C intake and mortality among a sample of the United States population. Epidemiology 1992; 3: 194–202. Galley HF, Thornton J, Howdle PD, Walker BE, Webster NR. Combination oral antioxidant supplementation reduces blood pressure. Clin Sci 1997; 9: 361–65. Taddei S, Virdis A, Ghiadoni L, Magagna A, Salvetti A. Vitamin C improves endothelium-dependent vasodilation by restoring nitric oxide activity in essential hypertension. Circulation 1998; 97: 2222–29.

Evans Department of Medicine and Whitaker Cardiovascular Institute, Boston University School of Medicine, Boston, MA 02118, USA (S J Duffy MRCP, N Gokce MD, M Holbrook MS , A Huang PhD, J F Keaney MD, J A Vita MD); and Linus Pauling Institute at Oregon State University, Corvallis, OR (B Frei PhD) Correspondence to: Dr Joseph A Vita, Cardiology, Boston Medical Center, 88 East Newton Street, Boston, MA 02118, USA (e-mail: [email protected])

Magnetic-resonance-guided percutaneous laser ablation of uterine fibroids P Law, W M W Gedroyc, L Regan Percutaneous laser ablation of uterine fibroids with magnetic resonance thermal monitoring causes shrinkage of treated areas by 37·5% 3 months later. This technique may provide an alternative to open surgery.

For the treatment of uterine fibroids, myomectomy and embolisation are the only options for women who want to retain their fertility. Both these choices have complications with relatively low post-procedural fertility rates.1–3 Endoscopic laser myolysis was first described in 19894 as a technique that could decrease fibroid size. Widespread use of thermo-ablative techniques in gynaecology have been limited by the inability accurately to predict lesion size leading to striking in-vivo variability and concerns over safety. Magnetic resonance imaging (MRI) provides real-time thermal imaging maps of the treated area and can be used to carry out real-time monitoring of thermal tissue alteration thereby overcoming many of these potential concerns. We describe an effective and accurate method for thermal ablation of uterine fibroids with an open MR interventional scanner to guide the percutaneous

THE LANCET • Vol 354 • December 11, 1999

Sequential images of fibroid during ablation over 20 min Monochrome images on the left show tissue changes seen without Real Time Image Processing. Those on the right show the colour changes seen as temperature within the fibroid increses to 55°C (green).

insertion of laser fibres and record the extent of uterine fibroid necrosis during treatment. 12 women with symptomatic fibroids who had not responded to medical management and who had completed their family were recruited to receive this procedure (Signa SPIO GE Medical Systems, Milwaukee, USA) under local anaesthetic. All women were awaiting hysterectomy, and were willing to undergo laser ablation before surgery. Ethics committee approval was obtained and laser ablation was carried out as a day case underlocal anesthetic. Four MRcompatible 18 gauge needles were placed within the target fibroid under MR guidance. After optimum needle positions were confirmed, bare laser fibres were placed directly into the centre of the fibroid via the outer needle sheath. All four sheaths were then pulled back by 2 cm to allow exposure of the bare fibre tip at the distal end of the needle. An infra-red diode (Diamed, Huntingdon, UK) laser power source was used with a four-way optical splitter with 5 watts of power being delivered per laser fibre. Tissue effects of thermal ablation were monitored with realtime image processing software which produced a real-time thermal map (figure). Thermal ablation was terminated when a maximum area of continual green colour was seen within the treated area; this represents a tissue temperature of greater than 55°.5 6 weeks after laser ablation, four women proceeded with their planned open surgery. Histological examination after surgery confirmed the presence of well defined necrotic areas measuring 3–4 cm in diameter, demonstrating the ability of this technique to cause coagulative necrosis in a controlled fashion with minimum damage to surrounding tissue. The remaining eight women declined their planned hysterectomy and underwent MRI to monitor fibroid shrinkage. 3 months after ablation, treated fibroid volume had decreased by 37·5% (range 25–49%). All women reported an improvement in symptoms of dysmenorrhoea, urinary frequency, and abdominal discomfort. Preliminary results suggest that this minimally invasive intervention may offer a realistic alternative to surgery with fewer complications and no in-patient stay. A substantial area of controlled fibroid destruction can be produced with an accurate correlation between the treated area seen on thermal imaging and histological examination. No endometrial damage related to thermal necrosis was seen, suggesting that fertility may not be affected by this procedure. Supported by a grant from the North West Thames NHS Research and Development Fund.

2049

1

Tulandi T, Murray C, Guralnick M. Adhesion formation and reproductive outcome after myomectomy and second look laparoscopy. Obstet Gynaecol 1993; 82: 213–15. 2 Goodwin S, Walker W. Uterine artery embolisation for the treatment of uterine fibroids. Curr Opin Obstet Gynaecol 1998; 10: 312–20. 3 Vashisht A, Studd J, Carey A, Burn P. Fatal septicaemia after fibroid embolisation. Lancet 1999; 354: 307–08. 4 Nisolle M, Smets M, et al. Laparoscopic myolysis with the neodymium-Ytrium Aluminium Garnet Laser. J Gynaecol Surg 1993; 9: 95–99. 5 Anzai Y, Lufkin R, Hirschowitz S, Farahani K, Castro DJ. MR imaginghistopathological correlation of thermal injuries induced with interstitial Neodymium-Ytrium Aluminium Garnet; YAG laser irradiation in the chronic model. J Mag Res Imag 1992; 2: 671–78.

Academic Department of Obstetrics and Gynaecology (P Law MB, Prof L Regan FRCOG), and Department of Interventional MRI (W W Gedroyc FRCP), ICSM@St Mary’s Hospital, London W2 1NY, UK Correspondence to: Dr P Law

15-month efficacy of maternal oral zidovudine to decrease vertical transmission of HIV-1 in breastfed African children DITRAME ANRS 049 Study Group* In a randomised, double-blind comparison of a maternal short regimen of oral zidovudine with placebo in West Africa the 15month efficacy of zidovudine in preventing mother-to-child transmission of HIV-1 infection was estimated at 30% (95% CI ⫺2 to 52).

Two randomised trials have shown that mother-to-child transmission of HIV-1 can be effectively decreased in breastfed African children with a short maternal regimen of oral zidovudine. 1,2 Because transmission of HIV-1 via breastfeeding can occur as long as exposure persists,3 there is concern that the benefit of the intervention may not be sustained in the long term. We report the results of the DITRAME ANRS 049a trial conducted in Abidjan, Côte d’Ivoire, and Bobo-Dioulasso, Burkina Faso, which assessed after 15 months of follow-up of breastfed children the efficacy of a peripartum zidovudine regimen. The methods of this phase II/III randomised double-blind placebo-controlled trial have been described previously.1 The oral regimen of zidovudine or matching placebo was: daily treatment of 500 mg or 600 mg starting at 36–38 weeks of gestation, a single oral loading dose of 500 mg or 600 mg at the beginning of labour, a 7-day postpartum treatment of 500 mg or 600 mg per day, and no treatment of the neonate. Clinical follow-up of and blood collection from the child were scheduled within 1 week of birth, then at day 45, day 90, and every 3 months until 15 months. Feeding practices were reported at each visit. The sample collected at day 180, or an earlier one when that was not available, was systematically processed by PCR. Time

Day 0 Day 45 Day 90 Day 180 Day 270 Day 365 Day 450

Children at risk Zidovudine

Placebo

193 159 144 141 102 71 51

198 147 132 126 92 59 46

If that sample was positive, PCR was then applied to all the preceding available samples. DNA PCR was systematically used in Abidjan. In Bobo Dioulasso, samples were analysed during phase II by DNA PCR and quantitative plasma RNA PCR (Amplicor HIV monitor, version 1.5, Roche Diagnostics Systems Inc, Branchburg, NJ, USA) and then by RNA PCR only. The diagnosis of HIV-1 infection in children was defined on the basis of one positive PCR. Serum samples collected at 9–15 months of age were screened for HIV-1 and HIV-2 antibodies by ELISA (Genelavia Mix, Diagnostics Pasteur, France, or Murex ICE 1-0-2, Murex Biotech Ltd, UK). Confirmation on the same sample was obtained with a synthetic peptide ELISA (Peptilav 1–2, Diagnostics Pasteur). A positive antibody test at 15 months was also a diagnostic criterion of HIV-1 infection in the child. Absence of infection was defined as a negative diagnosis 60 days or more after complete cessation of breastfeeding. A negative diagnosis on the last available sample while the child was still being breastfed or had been weaned for less than 60 days was defined as provisional absence of infection. Children who had no sample available for PCR and could not be followed up beyond 6 months of age were classified as of unknown HIV-1 status. Enrolment ended in February, 1998.1 The primary analysis of long-term efficacy was by intention-to-treat for liveborn children. Date of diagnosis of infection was the date of the first positive PCR test for children diagnosed before day 45 or the midpoint between the last negative and the first positive tests for those diagnosed after day 45. Endpoint date was 60 days after weaning for definitely uninfected children and the date of the last available negative test for children classified as provisionally uninfected. The probability of diagnosis by HIV-1 infection at a given age and the probability of breastfeeding were estimated by the Kaplan-Meier survival technique, and comparisons were made by the log-rank test. Efficacy was defined as 1 minus the ratio of the estimated transmission risks in the two treatment groups. Between September, 1995, and February, 1998, 431 women were enrolled; 401 livebirths were included in this analysis after random selection of one livebirth in six pairs of twins. The trial profile is similar to that in our previous report,1 with updated data. There were 201 liveborn infants in the placebo group and 200 in the zidovudine group, with 236 in Abidjan and 165 in Bobo Dioulasso. The median duration of the antepartum treatment was 21 days; intrapartum treatment was reported in 81% of deliveries and median length of postpartum maternal treatment was 7 days, without difference between the groups. Caesarean section was done in 2·5% of women. Nine children in the zidovudine group and four in the placebo group were bottle fed from birth (p=0·16). The probability of being breastfed at 6 months, calculated for the remaining 388 children, was 90·7% in the zidovudine group and 93·8% in the placebo group; that at 15 months was 45·6% and 40·1%, respectively, without difference between the treatment groups (p=0·56). Among the 401 livebirths, seven children in the zidovudine group and three in the

Cumulative number definitely uninfected

Cumulative number infected

Probability (%) of diagnosis of infection (95% CI)

Zidovudine

Placebo

Zidovudine

Placebo

Zidovudine

Placebo

·· 0 6 7 38 63 74

·· 0 3 4 22 51 61

·· 28 31 33 34 36 37

·· 42 48 52 57 57 57

·· 14·8 (9·6 to 20·0) 16·5 (11·1 to 21·9) 17·7 (12·1 to 23·3) 18·3 (12·6 to 24·0) 20·1 (14·0 to 26·2) 21·5 (14·9 to 28·1)

·· 21·7 (15·8 to 27·6) 25·1 (18·8 to 31·4) 27·4 (20·2 to 33·9) 30·6 (23·8 to 37·4) 30·6 (23·8 to 37·4) 30·6 (23·8 to 37·4)

% efficacy* (95% CI)

·· 32 (⫺5 to 56) 34 (1 to 56) 35 (5 to 56) 40 (13 to 59) 34 (5 to 55) 30 (⫺2 to 52)

Children of unknown HIV-1 infection status excluded. *1 minus the ratio of estimated transmission risks in the two groups.

Probability of HIV-1 infection diagnosis in 391 children during the first 15 months of life

2050

THE LANCET • Vol 354 • December 11, 1999