Ambulatory Blood Pressure Monitoring in PLHIV in Malawi: Preliminary Findings

To the Editors: The increase in survival of people living with HIV (PLHIV) and the economic and demographic changes in sub-Saharan Africa are accompanied by a rising number of patients suffering from cardiovascular disease.1,2 In particular, arterial hypertension represents a growing problem among PLHIV.3 According to a recently published meta-analysis, the prevalence of arterial hypertension rises from 10.5% in treatment-naive patients to 14.5% for those on antiretroviral therapy (ART).4 The prevalence of arterial hypertension in PLHIV from sub-Saharan Africa is estimated to range from 20% to 40%, corresponding to a total number of 5–10 million patients.3,5–7 Twenty-four-hour ambulatory blood pressure monitoring (24-h-ABPM) has been established as the most accurate noninvasive method for the diagnosis of hypertension.8,9 This technique can characterize circadian BP profiles, and nighttime BP has been shown to be superior to office BP (OBP) in predicting total and cardiovascular mortality.10 In addition, 24-h-ABPM is able to identify white coat hypertension (WCH), ie, isolated office hypertension.8,9,11,12 Although the importance of 24-h-ABPM had long been established in Europe,13 it became only recently acknowledged in the latest US guidelines.14 However, the availability of this technique in sub-Saharan Africa is limited. The prevalence of HIV infection in Malawi has been estimated at 9.6% in 2017,15 and 51.5% of the approximately 17 million people live below the poverty line of 1.90 US$/day.16 The present pilot study aimed at (1) testing the feasibility of 24-h-ABPM in a low-income country, (2) describing circadian BP profiles of PLHIV and HIV-uninfected (HIV−) controls, and (3) verifying the rate of true arterial hypertension among PLHIV. PLHIV with a medical history of hypertension from the ongoing prospective Lighthouse Tenofovir Cohort (LighTen) Study (ClinicalTrials.gov NCT02381275) that is conducted at the Lighthouse Clinic in urban Lilongwe, Malawi, were asked to participate in this subgroup analysis. HIV− employees of the Lighthouse Clinic served as controls. OBP was determined as the mean of the following measurements: (1) BP measurement as part of anthropometric measurements after entering the clinic, (2) additional measurement by a qualified nurse if BP > 140/90 mm Hg, and (3) manually induced measurement after positioning of the cuff for the 24-h-ABPM (Mobil-O-graph; IEM GMBH, Stolberg, Germany). All measurements were done in the sitting position after 5 minutes of rest with adjusted cuff sizes.9,17 A trained clinic employee, who was initially instructed and supported by an experienced investigator, explained the 24-h-ABPM procedure to patients and controls. Daytime (BP measurements every 15 minutes) and nighttime (BP measurements every 30 minutes) were originally defined from 6 am to 10 pm and from 10 pm to 6 am, respectively. However, as many participants forgot to use the designated button, daytime and nighttime were manually adjusted based on the most available markings (9 am–9 pm and 1 am–6 am, respectively). Normotension was defined as the mean OBP <140/90 mm Hg plus mean 24-h-BP <130/80 mm Hg plus mean daytime BP <135/85 mm Hg plus mean nighttime BP <120/70 mm Hg. Hypertension was defined as the mean OBP ≥140/90 mm Hg (≥2 measurements on ≥2 occasions), mean 24-h-BP ≥130/80 mm Hg, mean daytime BP ≥135/85 mm Hg, or mean nighttime BP ≥120/70 mm Hg. WCH was defined as OBP ≥140/90 mm Hg with normotension during all periods of 24-h-ABPM.13 Fifty-three PLHIV from the original LighTen study cohort with a history of hypertension were alive and on ART under active follow-up, and 39 of them could be reached through phone calls. Thirty-four nondiabetic patients consented to participate as well as 20 nondiabetic HIV− controls. Although information on salt consumption and physical activity was not available, all participants were nonsmokers with a comparable distribution of body mass index (BMI). There was no significant difference in the number of valid ABPM measurements for normotensive or hypertensive PLHIV and HIV− controls, and only one patient prematurely stopped the recording because of annoyance. OBP measurements and 24-h-ABPM revealed normal BP without antihypertensive therapy in 13 PLHIV, despite a history of hypertension. Nineteen of the remaining 20 PLHIV stated explicitly to take antihypertensive drugs as prescribed. According to the OBP definition, BP was controlled, that is, <140/90 mm Hg, in 5 of these 19 PLHIV (26.3%) and 2 of 8 treated hypertensive HIV− controls (25%). Irrespective of hypertension, a normal nocturnal BP reduction was seen in 9 HIV− controls (45%) and only 5 PLHIV (15%) (Table 1). Five of 20 PLHIV with office hypertension (25%) and one of the hypertensive HIV− controls fulfilled the criteria of WCH.TABLE 1.: Characteristics of HIV+ Patients and HIV–− ControlsThe findings from this pilot study confirm that 24-h-ABPM is feasible in resource-limited settings, such as Malawi, with some caveats regarding the reliable documentation of events during the recording period. In addition, the information from the medical history concerning hypertension seems to be considerably flawed in this study population. Compared with HIV-uninfected people, the risk for cardiovascular events is increased in PLHIV because of a complex interplay of traditional cardiovascular risk factors, HIV-specific factors, and concomitant metabolic disorders.18 Despite optimal suppression of virus replication, the ongoing systemic inflammation together with direct vascular-damaging effects of ART can induce endothelial dysfunction and promote the development of atherosclerotic lesions.18–20 Another involved factor could be a disturbed circadian BP pattern, which we also observed in our study, albeit the numbers are small, and need confirmation before definite conclusions. An abnormal nocturnal decrease in BP (“nondipping” = nighttime mean BP <10% lower than the daytime mean BP) has been found to be associated with a higher cardiovascular event rate.21 A recently published meta-analysis found the risk of nondipping to be on average 2.7 times higher in PLHIV compared with HIV− controls,22 and the mean nocturnal decrease in BP was lower in black compared with white HIV− participants from South Africa.23 A nondipping status was found in 52.9% of HIV− participants from the same geographical region,24 which is comparable with the rate found in the HIV− controls from this study. Predominantly, black South African female PLHIV on ART were included in 2 studies, and an abnormal dipping status was found in 24 of 30 (80%) and 43 of 67 (64%) patients.20,24 It is unclear whether abnormal BP profiles should change antihypertensive treatment decisions. However, in a large observational study, a higher number of patients with controlled BP, less nondipping, and a decreased cardiovascular event rate were seen if at least 1 antihypertensive drug was taken at bedtime.25 When an OBP of ≥140/90 mm Hg plus a normal mean value for the whole 24-hour period was applied, the prevalence of WCH was found to be 4%–8% in South Africa and 9% in a prospective Italian cohort study, with an increased prevalence of left ventricular hypertrophy.26,27 If instead only a normal daytime mean was used to define WCH, a prevalence of 12.2% was shown with a cardiovascular mortality 3 times as high as in patients with prehypertension.28 However, using the European Society of Hypertension criteria,13 the prevalence of WCH in untreated and treated hypertensive patients turned out to be 27.2% and 26.1%, respectively, and typical hypertension–mediated organ damage (left ventricular hypertrophy, microalbuminuria, or decreased glomerular filtration rate) was observed at the same rate as in normotensive study participants.29 In view of these findings from a registry including more than 100,000 patients, it seems justified to postpone antihypertensive drug treatment in this type of WCH. In studies from Spain (WCH definition applying a daytime mean of <135/85 mm Hg) and Norway (definition applying the daytime mean but <130/85 mm Hg), the prevalence of WCH in PLHIV amounted to 39%30 and 26%,31 respectively. To the best of our knowledge, no data on the prevalence of WCH in PLHIV from sub-Saharan Africa have been published so far. According to OBP criteria, hypertension was controlled in only one-fourth of PLHIV and HIV− controls despite antihypertensive therapy. Clearly, there is a need for improvement. However, among the study participants who had been diagnosed with hypertension but left untreated during former contacts with health care facilities, a substantial number turned out to be definitely normotensive. Measures of psychological distress were higher in mislabeled normotensives; they had a worse perception of their health and more symptoms of depression.32 When 24-h-ABPM is not available, it is mandatory to follow the rules when diagnosing arterial hypertension during office visits, that is, several BP measurements on several different occasions.14,17 This holds true, especially in PLHIV in whom the additional pill burden might endanger adherence to life-saving ART. In conclusion, 24-h-ABPM is feasible in resource-limited settings with some caveats regarding the reliable documentation of events during the recording period. A pathological nocturnal dipping pattern with its inherent unfavorable effects on cardiovascular risk was identified in most participating PLHIV from urban Malawi. In addition, WCH with probably a favorable cardiovascular prognosis was found in a subgroup of patients. These findings together with the frequently flawed medical history concerning hypertension warrant further prospective studies to confirm the clinical benefit of 24-h-ABPM, especially for PLHIV from sub-Saharan Africa.