Five Weeks Pregnant, and the Data That Finally Caught Up
The dolutegravir-pregnancy controversy that once made headlines has moved substantially since 2018 — the harder question now is how much residual caution five weeks of gestation still deserves, not whether the original alarm was fully justified.
C.N., a 29-year-old woman, found out she was pregnant and HIV-positive within the same week, at what her dating ultrasound confirmed was five weeks' gestation — news she describes taking in two very separate waves rather than as one combined shock. She works as a physical therapist at a rehab clinic and has been trying to conceive with her partner for about eight months; the pregnancy itself is very much wanted, which she says makes the HIV diagnosis land even harder, since she'd spent those eight months imagining a very different set of decisions to make right now. She has no prior HIV testing history, no other chronic illness, and reports no symptoms suggestive of acute retroviral illness at any point she can recall. Her partner has agreed to be tested this week as well, though his result isn't back yet and isn't part of today's decision.
Her CD4 count is 480 cells/µL and her viral load is 34,000 copies/mL. Five weeks' gestation places her inside the neural tube closure window, the exact period at the center of a real controversy in HIV-in-pregnancy care: the 2018 Tsepamo study first raised alarm about a possible link between periconception dolutegravir exposure and neural tube defects, prompting several guideline bodies to temporarily recommend against dolutegravir in early pregnancy. That same surveillance study's later, larger dataset substantially revised the estimate downward — current figures put the prevalence at roughly 0.19% with periconception dolutegravir exposure against a background rate near 0.11% — and IMPAACT VESTED subsequently found dolutegravir-based regimens outperformed efavirenz-based ones on viral suppression speed and maternal outcomes. Both WHO and DHHS have since revised their guidance to recommend dolutegravir as preferred in pregnancy, including at conception. The alarm that once shaped this exact conversation has been substantially, though the team agrees not completely, resolved by the data that came after it — and her own timeline, five weeks along with closure still roughly a week away, is exactly where that residual uncertainty still has room to matter.
Prenatal HIV clinic, first visit after diagnosis
The original 2018 Tsepamo signal is what most people still remember about this question, but it isn't the current data. The updated Tsepamo surveillance found the excess neural tube defect risk far smaller than first estimated, and IMPAACT VESTED found dolutegravir-based regimens achieved faster suppression and fewer adverse pregnancy outcomes than efavirenz-based regimens. Both WHO and DHHS now recommend dolutegravir as preferred in pregnancy, including at conception. I'd start it now.
I'm not disputing that the updated numbers are reassuring, and I'm not arguing for efavirenz for the rest of her pregnancy.
But 'far smaller than first feared' is still a real, nonzero signal, and she is currently inside the exact window that signal concerns — neural tube closure happens by roughly six weeks post-conception, and she's at five. Efavirenz has a much longer track record across exactly this window with no confirmed teratogenic signal in large cohorts. I'd start efavirenz now and switch her to dolutegravir once she's past the closure window, which keeps nearly all of its later advantages while sidestepping the one period where genuine, if small, uncertainty remains. Her CD4 of 480 gives us room to make that call calmly, too — she isn't someone whose immune status makes even a two-week delay to the faster-suppressing regimen feel urgent.
It might help to name the actual number both of you are really arguing about. The updated Tsepamo data puts periconception dolutegravir's neural tube defect prevalence at roughly 0.19%, against a background rate around 0.11% — an absolute difference of about 0.08 percentage points. Whether that's small enough to accept immediately or still large enough to defer past six weeks isn't really a data disagreement between you two anymore; it's a threshold question, and I think naming it plainly is more useful than either of you treating the other's position as uninformed by the same numbers.
Agreed: efavirenz-based therapy started today, with a planned switch to a dolutegravir-based regimen once she is confirmed past the neural tube closure window at her next visit. The team's disagreement resolved into a timing compromise rather than a genuine split — both positions accepted the updated data as reassuring, and differed only on how much residual caution five weeks of gestation still warranted.