Summary reports from this meeting include:

• HAART use among women in UK receiving treatment prior to conception

• Duration of ruptured membranes and vertical transmission in the UK

• Significant rates of unplanned pregnancies among young women born with HIV

• Route of HCV transmission in HIV-positive gay men is unlikely to be from semen

• 75% HIV-positive children have insufficient levels of Vitamin D

• Increase in LGV cases in gay men reported in the UK

• High rates of osteopenia and osteoporosis: importance of DEXA monitoring

• Summaries of other studies

Abstracts from the conference are published as a supplement to the May 2010 edition of HIV Medicine; Volume 11, Supplement 1.

Until these are posted on the aegis.org conference abstract database, a PDF files abstract is:

http://www.aegis.org/conferences/BHIVA/2010/16BHIVA-2010.pdf

HIV positive women in the UK are increasingly receiving HAART prior to conception and pregnancy.

BHIVA guidelines recommend that women already on treatment at conception remain on HAART throughout and after their pregnancy.

Some drugs, notably efavirenz (EFV), are not recommended for use during pregnancy.

Loveleen Bansi and colleagues from the UK CHIC Study and the National Study of HIV in Pregnancy and Childhood (NSHPC) analysed clinical and treatment patterns of women conceiving their first child on HAART between 1996 and 2008 using linked data from the two datasets.

The investigators found 1838 matches between the women in the current UK CHIC (n=8,659) and NSHPC (10,912) datasets. Of these, 821 (45%) had received HIV clinical care before their first pregnancy.

The majority of women were infected via heterosexual sex (88%) and over two-thirds (69%) were of black-African ethnicity. Their median age at delivery was 33 (IQR: 28-36) years.

Just over half, 440/821 (54%) women were receiving HAART at the time of conception of their first child.

Their median CD4 count at conception was 389 (IQR 270-554) cells/mm3. Amongst women who had a measurement up to 90 days before conception, 88 (27.9%) had a detectable viral load >50copies/mL.

Of the 440 women, 237 (53.9%) received an NNRTI and for 86 (19.5%) women this was EFV. Most women had not started treatment close to conception, 40.9% had already been on HAART for over 3 years and only 10.9% started less than 6 months before conceiving.

One-third (n=155 (35.2%)) made a switch in their regimen before delivery. The proportions of women switching therapy by 3 and 6 months of conception were 22% and 33%. Of those receiving EFV at conception, 37 (43%) of women switched this drug.

The vast majority (97%) of women receiving HAART at conception were also receiving HAART at delivery. After delivery 286/428 women switched regimen at a median of 15 (IQR 13-18) months.

The proportions of women switching regimen at 6, 12, 18 and 24 months were 27%, 41%, 55% and 61%, respectively. In the year following delivery 13 (13%) of women receiving HAART at delivery discontinued completely.

The investigators wrote; “ Adherence support is important after pregnancy to minimise the number who interrupt or stop treatment after delivery.”

comment

The authors of this poster noted (personal correspondence) that they don’t have information on when the woman found out she was pregnant, in relation to conception (or when she told her clinician).  So, for women who were already on efavirenz at the time of conception, if they didn’t find out they were pregnant until a few months into the pregnancy, it might be argued that any potential damage had already been done, and so there was less reason to switch.

The study also highlights the importance of post natal adherence support.

Ref: Bansi L et al. Use of antiretroviral therapy during and after pregnancy among HIV-infected women already aware of their infection before conceiving. 2nd Joint Conference of BHIVA with BASHH, 20–23 April 2010, Manchester. Poster abstract P154.

http://www.aegis.org/conferences/BHIVA/2010/16BHIVA-2010.pdf

Longer duration of rupture of membranes (ROM) was identified as a risk factor for mother to child transmission (MTCT) in the 1990s. Elective caesarean section prior to ROM was found to be protective.  However, these studies were conducted before the availability of HAART.

Whether duration of ROM has clinical implications for women on effective HAART is unclear. An increasing number of women in the UK opt to deliver vaginally but, of these, a high proportion, undergo emergency caesarean section. It is likely that concern over ROM contributes to this management decision.

In an oral presentation at the 2010 BHIVA/BASHH meeting, Pat Tookey showed findings from an investigation, using routine surveillance data from UK and Ireland, to explore the association between duration of ROM and transmission. Surveillance of obstetric and paediatric HIV is conducted through the National Study of HIV in Pregnancy and Childhood (NSHPC). Data on ROM have been collected since January 2007.

In this study, the investigators reviewed pregnancies resulting in live singleton births among HIV-positive women reported in 2007-9.

During this period, 2686 births were reported. The majority (95%) of mothers were on HAART; 40% had an elective caesarean section, 34 had vaginal delivery and 26% emergency caesarean section. Almost Three quarters of mothers (74%) had an undetectable viral load and the rate of mother to child transmission was 0.9% (15/1697).

Of the total, 1063/2686 mothers had an elective caesarean and data were missing for 298. There were ROM data for 1325, of which 884 (67%) had ROM prior to delivery and data on duration was provided for 661 (75%). The median duration was 4 hrs (IQR 1.5-8), <6hrs for 444 (67%), >6-48 hrs for 217 (33%), this included 16 mother with >48 hrs ROM.

There were 6/421 (1.4%) transmissions overall, among the infants with confirmed HIV status. The rate was similar for infants with < 6 hrs ROM, 2/284 (1.4 compared to those with  >6 hours, 2/137 (1.5%), OR 1.0 (95%CI 0.2-5.7, p=1.0)

In the sub group of women with undetectable viral load (<50 copies/mL) near delivery (99.7% on HAART), there was no difference in MTCT (overall 1/341, 0.3%), between those with ROM for >6 hrs compared to <6 hrs (0/112, 0.4% vs 1/229, 0.0%, p=1.00). Likewise, among mothers with undetectable viral load who had a planned vaginal delivery (overall 1/203, 0.5%:  0/52, 0% vs 1/151, p=1.0).

Six of the 661 children had confirmed infection at the time of analysis including 3 likely in utero transmissions (positive PCR on Day 1, see Table 1).

Table 1: Six children infected with HIV who were born to women with ROM before delivery

* closest to delivery. Note: Patients 1, 5 and 6 had reported adherence issues.

The investigators concluded that although data are sparse to date, so far there is no evidence that among women on HAART longer duration of ROM is associated with an increased risk of MTCT. Larger sample sizes are required and comprehensive data on: ROM and duration of ROM, infant infection status and viral load close to delivery. Continuing monitoring is essential.

Ref: Haile-Selassie H et al. Duration of ruptured membranes and vertical transmission of HIV: data from national surveillance in the UK and Ireland. 2nd Joint Conference of BHIVA with BASHH, 20–23 April 2010,  Manchester. Oral abstract O2.

http://www.aegis.org/conferences/BHIVA/2010/16BHIVA-2010.pdf

There is little research and information reported on young women growing up with HIV. But a study at the meeting showed some interesting data on the pregnancy outcomes of young women who were born with HIV 10–20 years ago. The study looked at a cohort of 172 women, born with HIV and accessing services in 19 centres across the UK and Ireland.

Overall, there were 36 pregnancies reported in 27 women, with a median age of 18 years at first pregnancy. Of these, 27 (75%) were unplanned, 7 (19%) were planned and 2 were unknown. The women all conceived before September 2009. Seventeen women were on HAART, with a median CD4 count of 244 cells/mm3, but 8 out of the10 women who were not on treatment had CD4 counts below 200 cells/mm3.

Of the reported pregnancies, 86% (31/36) were with regular partners, of whom 22 (61%) reported being unaware of their partners HIV status. The pregnancies resulted in 5 (14%) 1st trimester miscarriages, 9 (25%) elective terminations, 18 (50%) live births and 4 (11%) pregnancies were ongoing.

At the time of delivery, 89% of the mothers were on HAART, with a median CD4 count of 254 cells/mm3 and a median viral load of 79 copies/mL (range <50–580,000). Seven women delivered with viral load <50 copies/mL, four had VL >1,000 copies/mL. Two women were admitted for Directly Observed Therapy and two were non-adherent to HAART at delivery. Mode of delivery was 9 elective and 5 emergency C-sections with 4 vaginal deliveries. 6 (33%) infants delivered at <37/40, five of whom required Neonatal Intensive Care. None of the babies were infected and there were no congenital anomalies were reported. Five of the babies were fostered and 3 have ongoing concerns.

The study showed without a doubt, that prevention of mother to child transmission worked, even though some mothers had advanced HIV and were not well adherent. Additionally, the study also highlighted a worrying high number of unplanned pregnancies, a population with social needs despite access to contraceptive services.

Ref: Williams B et al.  Pregnancy outcomes in women growing up with HIV acquired perinatally or in early childhood. 2nd Joint Conference of BHIVA with BASHH, 20–23 April 2010, Manchester. Poster abstract P144.

http://www.aegis.org/conferences/BHIVA/2010/16BHIVA-2010.pdf

The mechanism for high rates of hepatitis C (HCV) transmission in HIV-positive gay men is unknown with little data on whether HCV levels in semen had a similar risk factor to HIV viral load, especially during acute infection.

Joanna Turner and colleagues presented a new study to inform this field. [1]

Paired blood and semen samples were collected from 5 acute and 9 chronic HCV cases in HIV-positive men. At baseline 0/5 acute and 2/9 chronic cases had detectable HCV RNA in semen. Of all samples tested 2/10 (20%) of acute cases and 4/23 (17%) of chronic cases (p=NS) had detectable HCV RNA in semen.

However, when detected, HCV RNA viral loads were low: <30 IU/mL (acute cases) and <230 IU/mL (chronic cases) and did not correlate with plasma HCV viral load

Taken together, this lead the researcher to suggests that the quantity of seminal HCV virus is not a significant factor in determining the rate of HCV transmission, even during acute infection. Recruitment to the study is ongoing.

Comment

HCV levels in semen have been previously reported. [2] Results from the larger study will be important to understand whether infection from semen is likely to occur at these low levels.

References

1. Turner J et al. Hepatitis C viral load in semen of HIV-positive men during acute and chronic hepatitis C infection. 2nd Joint Conference of BHIVA with BASHH, 20–23 April 2010, Manchester. Oral abstract O5.

http://www.aegis.org/conferences/BHIVA/2010/16BHIVA-2010.pdf

2. See this earlier report in HTB from 2003.

Although the interpretation of the role HIV and vitamin D deficiency remains a focus in HIV-positive populations in relation to reductions in bone mineral density in adults, the impact on children is potentially more worrying.  If HIV-related complications prevent optimum bone development earlier in life (generally until the age of 30), this could result in higher rates of bone complication in later in life.

Atkinson and colleagues presented results of such low vitamin D levels in HIV-positive children that suggest further research should be an urgent priority.

An audit of plasma bone biochemistry, 25(OH) vitamin D and PTH levels in a cohort of 131 HIV-positive children receiving routine clinical care at a single UK centre between January and December 2009.  Median age was 12 years (IQR 9, 15); 51% were female and 85% were African/Caribbean. Median CD4 count (%) was 760 (32%) and 104 children (79%) were on HAART.

64 children (49%) were deficient (defined as 25nmol/L) and a further 37 (28%) had insufficient levels (25-50nmol/L). Abnormal PTH (>6.8 pmol/L) was seen in 15/52 children who had these levels (28.9%).

In multivariate analysis 25(OH) vitamin D deficiency was associated with older age (p=0.001), African/Caribbean ethnicity (p=0.04), winter season (0.008) and NNRTI use (P=0.01).

The authors concluded that Vitamin D deficiency and insufficiency is very common in children with HIV. Maximising bone health is increasingly important as this population enter adult life and the role of vitamin D supplementation requires further elucidation.

Ref: Atkinson S et al. Vitamin D deficiency in children with perinatally acquired HIV-1 infection living in the UK. 2nd Joint Conference of BHIVA with BASHH, 20–23 April 2010, Manchester. Poster abstract P159.

http://www.aegis.org/conferences/BHIVA/2010/16BHIVA-2010.pdf

Bone disease was addressed in many of the posters, with the first of these reports coming from Guys and St Thomas’ and the remaining three from the Chelsea and Westminster Hospital.

Perry and colleagues from reported results of a cross-sectional study of 175 randomly selected HIV-positive patients who completed lifestyle and general health questionnaires that were compiled with biochemical analyses and DEXA lumber spine and hip. [1]

Baseline characteristics included median age 38 years (IQR 30-43); 64% male, 41% black, 85% ARV-experienced, 31% current smokers.

DEXA results showed 49% patients had reduced BMD, 13% with osteoporosis and 36% with osteopenia. Age increased the association (p=0.007) occurring at a median age of 44.50 years (IQR 38–51 years) but not gender, with osteoporosis diagnosed in approximately 10% of patients aged 40-49 and 20% in those aged >50 years. (See Table 1).

Table 1: BDM results by gender and age

In multivariate analysis, other risk factors were low BMI (OR 0.87; 95%CI 0.79–0.95; p= 0.003) and ever having been on HAART (OR 4.43; 95%CI 1.57–12.50; p= 0.005). Gender, ethnicity, HIV viral load, CD4 cell count, CD4 cell count nadir and vitamin D were not statistically associated with abnormal BMD.

In their conclusion the authors highlighted the high proportion of patients with HIV from young age groups, and a significant correlation with HAART and that this may provide a rationale for routine screening for risk factors that predict fracture in HIV, including low BMD.

Stuart-Buttle and colleagues reported results from a retrospective audit of 106 patients who had DEXA scans from 2007–2009. [2]

12% had osteoporosis, 30% had osteopaenia and 58% had normal DEXA scans. Of the 44 patients over 50 (mean age 58.1 +7.02), 36% had a diagnosis of osteoporosis and 41% had osteopaenia, compared to 28% and 5%, respectively, in the group under 50 years.

While this was a retrospective study, presumably in patients selected for DEXA bone concerns, the researchers concluded that HIV should be considered a risk factor and that including HIV-positive people >50 years need to be included in screening studies. No significant correlation between bone mineral density and CD4 count, calcium or vitamin D levels.

Hughes and colleagues presented results from a new ‘over 50s clinic’ that was set up in January 2009. Of 54/70 patients with DEXA results (4 patients were excluded due to with diagnosed bone disease), osteopaenia was diagnosed in 24% (13/54) and osteoporosis in 11% (6/54). Of these, 77% (10/13) with osteopaenia and 100% (6/6) with osteoporosis were male. [3]

The mean age was 60 years, 93% (50/54) were male and 85% (46/54) of white ethnicity. All patients were taking ART (100% VL <50, mean CD4 551 cells/mm3).

Low vitamin D levels occurred in 66% (4/6) with osteoporosis, 38% (5/13) with osteopaenia and 49% (17/35) with normal DEXA result. With over one third of this cohort with osteopaenia or osteoporosis, the authors concluded that this supported routine screening in individuals aged over 50.

Finally, Rashid and colleagues, prospectively measured Vitamin D levels (25(OH) vitamin D) in 312 patients in July 2009. [4]

Mean age was 48 years (range 25-83), 88% male, mean duration of HIV infection 12 years (range 0-26). Median vitamin D level was 66nmol/L (range <10-221) with 35% levels low (40-70nmol/L) and 21% deficient (<40nmol/L). Low Vitamin D correlated with non-caucasian ethnicity (p<0.001) and female sex (p<0.001), but not antiretroviral class or specific agent, including efavirenz.

Of note, in 102 patients (33%) who had DEXA scans for unrelated reasons, median vitamin D levels were 71, 71 and 58 nmol/L for those with normal, osteopaenic and osteoporotic results respectively.

The lack of association of low vitamin D levels with DEXA results, and also with alkaline phosphatase levels, suggest the importance of larger definitive studies to inform patient management in this area.

References

All references are to the Abstracts of the 2nd Joint conference of the BHIVA and BASHH, 20–23 April 2010, Manchester.

Abstracts from the conference are published as a supplement to the May 2010 edition of HIV Medicine; Volume 11, Supplement 1.

http://www.aegis.org/conferences/BHIVA/2010/16BHIVA-2010.pdf

1. Perry M et al. The relationship of HIV and bone density: implications for screening. Poster abstract P44.

2. Stuart-Buttle  C et al. Screening for bone disease in HIV patients. Poster abstract P46.

3. Hughes A et al. Over 50? It’s time for a dual energy x-ray absorptiometry (DEXA) scan. Poster abstract P39.

4. Rashid T et al. No association of vitamin D levels with individual antiretroviral agents, duration of HIV infection, alkaline phosphatase levels or bone mineral density findings. Poster abstract P45.

Many other studies at the conference deserve further reporting but can only be briefly summariesed here. For full details please refer to the conference abstracts and contact the lead authors.

All references are to the Abstracts of the 2nd Joint conference of the BHIVA and BASHH, 20–23 April 2010, Manchester.

Abstracts from the conference are published as a supplement to the May 2010 edition of HIV Medicine; Volume 11, Supplement 1.

http://www.aegis.org/conferences/BHIVA/2010/16BHIVA-2010.pdf

Transmission and late diagnosis in older people: half of late diagnoses in people over 50 years old

An oral presentation from the Health Protection Agency highlighted some aspects of how older people are affected by HIV. The number of older adults who are HIV-positive in the UK from 2333 in 2000 to 8268 in 2007, accounting for 16% of adults accessing care in 2007 and 8% of all HIV diagnoses between 2000–2007. [1]

Compared to younger adults, newly diagnosed adults aged 50 years and over were more likely to be men (74% vs. 58%; p<0.001), infected through sex between men (40% vs. 34%; p<0.001) and of white ethnicity (60% vs. 38%; p<0.001). Older heterosexuals adults were more likely to be infected within the UK (16% vs. 12%; p<0.001), with evidence of travel abroad amongst white heterosexual men.

Late diagnosis (CD4 count <200) was significantly higher amongst older adults (48% vs. 33%; p<0.001); with older MSM being twice as likely to present late than younger MSM.

This study estimated that nearly half (48%; 1486) of persons diagnosed between 2000 and 2007 acquired their infection aged 50 and over.

Ref: Smith R et al. Refocusing our efforts – transmission and late diagnosis of HIV among adults aged 50 and over. Oral abstract O3.

Ocular syphilis at first presentation of HIV

Three cases were described where ocular syphilis was the presenting symptom: a 33-year old heterosexual man, a 20-year old gay man, and a 39-year old gay man. The study concluded: “Syphilis should be excluded in cases of uveitis and optic neuritis; other features of secondary syphilis may be absent. All patients had improving visual symptoms after neurosyphilis therapy, and had preceding oral steroids to prevent Jarisch-Herxheimer reaction, as this can worsen ocular symptoms. Early diagnosis is important as ocular syphilis can rapidly cause blindness.

Ref: Dhairyawan R et al. Ocular syphilis as the first presentation of HIV infection. Poster abstract P132.

Perinatally infection diagnosed in late adolescence

Two cases of extremely late diagnosis, in a young man and woman, both 20-year old Ugandan patients, presenting with multiple complications, including HIV-related dementia. These rare cases highlight the importance of family history and HIV testing in children and young adults who were potentially exposed to HIV during pregnancy and at birth.

Ref: Ross S et al. Vertical HIV infection in young adults presenting with HIV-associated dementia. Poster abstract P169.