Potential impact of new WHO pregnancy guidance (2010)

Polly Clayden, HIV i-Base

A research letter in the June 1 2010 issue of AIDS, authored by Louise Kuhn and colleagues described an evaluation of the potential impact of the WHO 2010 guidelines for initiating antiretroviral treatment in pregnant women.

The guidelines now recommend treatment for adults with stage 3 or 4 (irrespective of CD4 count), or with CD4 count of 350cells/mm3 and below (irrespective of clinical stage). Previously WHO guidelines required stage 3 if CD4 count was 200-350 cells/mm3.

The study was performed using data from 1025 HIV-positive women and infants followed for 24 months in Lusaka Zambia before the widespread use of ART. Children were breast fed for 4 months after which some women weaned their infants as randomised and some continued to breastfeed. Overall the median duration of breastfeeding was 12 months.

The investigators evaluated the associations between maternal characteristics measured during pregnancy, including CD4 count, viral load, clinical stage, and the old and new WHO treatment criteria.

They looked at the capacity of these factors to predict: maternal mortality between delivery and 24 months, and perinatal (detected before 6 weeks) and postnatal (detected after 6 weeks) transmission.

They show a detailed analysis of the percentage of the pregnant population who would require treatment, rates and relative risks of maternal mortality and perinatal/postnatal transmission. Assuming a fully effective intervention, they estimated the preventable proportion according to various criteria for initiating antiretroviral treatment.

They reported that in their cohort 54% of women had CD4 counts below 350 cells/mm3 and the majority of maternal deaths (88%) occurred in this group, RR 7.2 (95% CI, 3.6-14.5). Sixty-eight percent were eligible for treatment using the new criteria, which includes clinical stage 3 as well as CD4 count below 350 cells/mm3, and 92% deaths occurred in this group, RR 6.2 (2.7-14.2).

They explained that if the new criteria were applied, 10.1 women would need to be treated per death averted, using only CD4 criteria of less than 350 cells/mm3 the number of women needed to be treated would be 8.4.

When they looked at the effect of viral load criteria they found having a viral load of 48,428 copies/mL classified the same proportion of women needing treatment as a CD4 threshold of 350 cells/mm3 and below but only identified 76% of deaths.

If viral load was added to CD4 count they found this performed similarly to the new WHO criteria. Adding viral load to the new criteria would identify the maximum number of deaths (96%) but 76% of women would need to be treated.

They noted that although it has suggested that lower CD4 thresholds could be used in pregnancy to accommodate the effects of haemodilution, their data suggest that CD4 of 200, 250 and 350 cells/mm3 would only identify 59%, 72% and 79% of deaths respectively. Therefore they recommend that, “making this adjustment is unwise”.

Additionally the new criteria would detect 88% of perinatal and post natal HIV transmissions. Using CD4 criteria alone would detect almost as many postnatal transmissions (83%) but fewer perinatal (76%).

Viral load and CD4 are independent predictors of transmission. Multivariate analysis revealed viral load, RR 3.1 (95% CI, 2.0-4.6) to be a stronger predictor for perinatal transmission controlling for CD4, RR 2.0 (95% CI 1.3-3.0). For postnatal transmission they were similarly predictive: viral load RR 3.8 (95% CI 2.2-6.3) and CD4 RR 3.8 (95% CI 2.1-6.8). The investigators suggested: “Combining viral load and CD4 count as either/or criteria for initiating therapy would lead to better results than the new WHO criteria while treating slightly fewer women.”

If they adjust for ART to reduce transmissions effectively, treating women according to the new WHO criteria could prevent 82% of all infections even if no extended postnatal interventions are used among women not indicated for ART. They suggest that although it is desirable that these interventions are implemented they will be more costly per infection prevented, as the transmission rate among healthier women is lower.

The investigators write that their data suggest that the inclusion of clinical staging in the new criteria increases the number of women treated but with only marginal increase in coverage of women and children at risk. Although they acknowledge that it is useful in settings where laboratory testing cannot be done or is unreliable.

They acknowledge that estimates of the proportion of deaths and transmissions averted will vary across settings but note that the characteristics they observed are similar to those in a large multisite dataset from women participating in the MTCTPlus programme.

They concluded that their data provide evidence-based support for the thresholds in the revised treatment guidelines. They write: “Our analysis also provides estimates of the large positive impact these guidelines could have if widely implemented on reducing mortality among women and HIV transmission to children.”


This elegant analysis once again makes the case for treatment of pregnant women at 350 cells/mm3 both for their own health and prevention of transmission.

Kesho Bora data at IAS this year also supports this.

Ref: Kuhn L et al. Potential impact of new WHO criteria for antiretroviral treatment for prevention of mother to child HIV transmission. AIDS. Volume 24. Number 9. June 2010.

Links to other websites are current at date of posting but not maintained.