ABSTRACT
Accurate characterization of blood pressure (BP) phenotypes is essential for cardiovascular risk assessment, particularly among type 2 diabetes mellitus (T2DM) patients. We aimed to evaluate the prevalence of different BP phenotypes using multiple BP assessment methods, in Black Africans, with and without T2DM. Recently diagnosed T2DM patients (n = 151) and age‐ and sex‐ matched controls (1:1) without T2DM had their BP assessed using standard office BP measurement (OBPM), unattended automated office BP measurement (u‐AOBP), 24 h ambulatory BP monitoring (ABPM), and home BP monitoring (HBPM). Four BP phenotypes were defined, combining OBPM with each of the remaining methods: normotension, white‐coat, masked, and sustained hypertension. Using OBPM + ABPM, T2DM patients were less often classified as normotensive (32.5% vs. 48.3%) and more often as having masked hypertension (23.2% vs. 9.9%) than controls (p = 0.004). Both OBPM + HBPM and OBPM + u‐AOBP underestimated masked hypertension and overestimated with‐coat hypertension. When compared with OBPM + ABPM, concordance in the classification of participants into four phenotypes (κ, 95%CI) was higher for OBPM + HBPM (T2DM: 0.62, 0.52 – 0.71; controls: 0.70, 0.60–0.80) than for OBPM + u‐AOBP (T2DM: 0.51, 0.40–0.61; controls: 0.61, 0.50–0.72) and lower among participants with T2DM. Recently diagnosed T2DM patients showed a higher prevalence of masked hypertension than participants without T2DM. When combined with OBPM, HBPM and u‐AOBP failed to identify a large proportion of patients with masked hypertension and overestimated white‐coat hypertensions, highlighting the need to incorporate ABPM into evaluation protocols.
Keywords: Africa south of the Sahara, ambulatory, blood pressure monitoring, diabetes mellitus, hypertension, masked hypertension, type 2, white coat
1. Introduction
Accurate assessment of blood pressure (BP) is essential for the diagnosis, treatment, and control of hypertension, as well as for cardiovascular risk stratification [1]. Four BP phenotypes may be considered: normotension, white‐coat hypertension, masked hypertension and sustained hypertension [2], each carrying distinct prognostic implications for cardiovascular outcomes and guiding individualized clinical management strategies [3, 4].
Several methods are currently available for BP assessment. Standard office BP measurement (OBPM) remains the most common approach in clinical practice, although it is considered the least reliable [5]. Current international guidelines recommend confirming elevated OBPM values using out‐of‐office methods, including 24 h ambulatory BP monitoring (ABPM) or home BP monitoring (HBPM) [6, 7, 8]. Unattended automated office BP measurement (u‐AOBP), although still an office‐based technique rather than an out‐of‐office method, has been proposed as a standardized approach to reduce the white‐coat effect and improve the accuracy of office BP assessment [9, 10, 11]. All these methods are applied in both clinical practice and research, providing complementary approaches to the evaluation of BP [12].
T2DM is a major and rapidly growing global health challenge, particularly in Africa [13, 14]; in Mozambique, the prevalence increased from 2.9% in 2005 [15] to 7.4% in 2015 [16]. The prevalence of hypertension ranges from 50% to 80% in patients with type 2 diabetes mellitus (T2DM), being twice as high as in age‐matched individuals without T2DM [17, 18, 19, 20]. Particularly, masked hypertension has also been reported to be more common in individuals with T2DM, with prevalence varying between 13% and 66% [21, 22, 23]. The coexistence of T2DM and hypertension significantly increases the risk of macrovascular (stroke, myocardial infarction, peripheral arterial disease) and microvascular complications (renal failure and retinopathy), as well as other associated conditions such as sexual dysfunction [24, 25, 26]. Furthermore, individuals with T2DM often present greater BP variability, abnormal nocturnal patterns, and higher overall cardiovascular risk compared with individuals without T2DM [27]. This underscores the importance of using appropriate BP measurement methods when evaluating individuals with diabetes.
This study aimed to assess the prevalence of the four distinct BP phenotypes, using multiple BP assessment methods, in recently diagnosed T2DM patients and age‐ and sex‐matched controls without T2DM.
2. Methods
This study took place in two cities located in southern Mozambique: Maputo, the capital city, and Matola, the country's second‐largest city, each having a commuting population of over one million inhabitants [28, 29]. These cities share economic and social dynamics, and face common health challenges that require coordinated and integrated responses [30, 31].
2.1. Selection of Participants
The study population included Black African adults (≥ 18 years) residing in Maputo and Matola, not under antihypertensive treatment. For the purposes of this study, we recruited recently diagnosed T2DM, defined as a confirmed diagnosis within the previous three months before enrolment. T2DM was defined by fasting plasma glucose ≥ 7.0 mmol/L or HbA1c ≥ 6.5% [32]. Individuals with lower values at assessment were also included if previously diagnosed within three months and receiving antidiabetic treatment.
T2DM participants were recruited from selected health facilities in Maputo and Matola, as well as at the Mozambican Diabetes Association (AMODIA), in Maputo. Eligible individuals were identified by healthcare professionals and consecutively invited. Those who agreed were referred to the research centre at Maputo Central Hospital, where eligibility was confirmed and study procedures were conducted.
For each T2DM participant, one control without T2DM was recruited, matched by sex and age (±5 years), with no history of T2DM or antihypertensive treatment. Controls were mainly relatives of participants or recruited through other participants when necessary. Glycaemic status was biochemically confirmed.
2.2. Study Population
A total of 179 individuals with recently diagnosed T2DM were invited. Of these, 13 did not attend and 15 were excluded due to antihypertensive treatment. Among 177 initially invited controls, 23 were excluded because they were under antihypertensive treatment, and three did not complete the study procedures, leaving 151 eligible individuals. After these exclusions, 302 participants (151 patients with T2DM and 151 controls) were included in the study.
2.3. Participants’ Evaluation
BP was assessed in all participants using OBPM and complementary methods, namely u‐AOBP, ABPM, and HBPM. OBPM was performed at the first visit after completion of the questionnaire. Three consecutive readings were obtained at one‐minute intervals using an automated sphygmomanometer (Omron MIT5 Connect, HEM‐7280T‐E; Omron Healthcare Co., Ltd., Kyoto, Japan). An appropriate sized arm cuff of 22 to 42 cm circumference was used for all participants and measurements were performed on the non‐dominant uncovered arm, with participants seated calmly and the arm supported at heart level. The mean of the second and third readings was used for analysis, following current hypertension management guidelines [8, 33].
For u‐AOBP, an automated BP monitor (Omron HEM‐907‐E7; Omron Healthcare Co., Ltd., Kyoto, Japan) was programmed to record three consecutive readings at one‐minute intervals, after a five‐minute rest period, while the participant remained alone in a quiet room without interaction or distractions. Cuff size was selected according to upper‐arm circumference, using a medium cuff for arm circumferences of 22–32 cm and a large/extra‐large cuff for arm circumferences of 32–42 cm. This assessment was conducted immediately after the OBPM. The device automatically calculated the mean value using the three readings.
On the following day, ABPM was initiated using a fully automated device (OnTrack 90227; Spacelabs Healthcare, Snoqualmie, WA, USA). Appropriately sized cuffs (adult, 24–32 cm; large adult, 32–42 cm; extra‐large adult, 38–50 cm) were selected according to participants’ arm circumference, and BP was recorded over a 24 h period, at intervals of 20 min during the daytime and 30 min at nighttime. Participants were instructed to maintain usual activities and sleep routines, while keeping the arm relaxed during measurements [2]. Recordings were considered valid if at least 80% of measurements were successful and included at least one valid reading per hour; otherwise, monitoring was repeated. The 24 h ambulatory BP recordings were divided into daytime (06:00–22:00) and nighttime (22:00–06:00) periods [8] based on a standard time to define day and night for all participants.
After returning the ABPM device, participants were provided with a sphygmomanometer for HBPM and instructed to start the assessment the following morning. HBPM was conducted using the same device as OBPM. Participants were trained and received written instructions based on the European Society of Hypertension guidelines [8, 34]. They were asked to obtain three consecutive readings at one‐minute intervals, both in the mornings (07:00–09:00) and evenings (18:00–20:00) over seven consecutive days. The first day was excluded from analysis and the mean of the remaining six days was used [2, 8].
BP phenotypes were defined using OBPM as the reference, in combination with each of the complementary methods (u‐AOBP, 24 h ABPM or HBPM). Normotension was defined as normal office BP (systolic BP (SBP) < 140 mmHg and diastolic BP (DBP) < 90 mmHg) and a normal value on the complementary method used (SBP < 140 mmHg and DBP < 90 mmHg for u‐AOBP, SBP < 130 mmHg and DBP < 80 mmHg for 24 h ABPM, and SBP < 135 mmHg and DBP < 85 mmHg for HBPM). Masked hypertension was defined as normal office BP and elevated values on the complementary method (SBP ≥ 140 mmHg or DBP ≥ 90 mmHg for u‐AOBP, SBP ≥ 130 mmHg or DBP ≥ 80 mmHg for 24 h ABPM, and SBP ≥ 135 mmHg or DBP ≥ 85 mmHg for HBPM). White‐coat hypertension was defined as elevated office BP (SBP ≥ 140 mmHg or DBP ≥ 90 mmHg) and normal values on the complementary method. Finally, sustained hypertension was defined as elevated office BP (SBP ≥ 140 or DBP ≥ 90 mmHg) and elevated BP values on the complementary method [2, 3, 8, 35, 36].
Participants identified with elevated BP requiring clinical evaluation, including those with previously unrecognized hypertension and those with known but untreated hypertension, were informed of their results and referred to their treating physician or the nearest health‐care facility for further evaluation and management.
We computed the difference between the mean daytime and nighttime SBP values obtained by ABPM to determine the dipping pattern. Participants were classified as dipper (10%–20% night SBP fall), extreme dipper (> 20% night SBP fall), non‐dipper (< 10% night SBP fall), and reverse dipper (nighttime SBP higher than daytime SBP) [23].
Blood samples (approximately 6 mL) were collected to assess fasting glycaemia, lipid profile (total cholesterol, triglycerides, HDL and LDL), renal function (urea and creatinine), electrolytes (sodium and potassium), and HIV status.
Body weight and height were measured with participants wearing light clothing and no shoes and Body Mass Index (BMI) was calculated as weight in kilograms divided by the square of height in meters (kg/m2). BMI was further categorized in underweight (BMI < 18.5 kg/m2), normal weight (BMI 18.5–24.9 kg/m2), overweight (BMI 25.0–29.9 kg/m2), and obesity (BMI ≥ 30.0 kg/m2) [37].
Waist circumference was measured at the midpoint between the lower margin of the last palpable rib and the top of the iliac crest, and hip circumference at the widest portion of the buttocks [38]. The waist‐to‐hip ratio (WHR) was calculated by dividing waist circumference by hip circumference. WHR is an anthropometric indicator of central adiposity and cardiometabolic risk. Participants were classified into low‐, moderate‐ and high‐risk categories using sex‐specific WHR cut‐offs (women: ≤ 0.80, 0.81–0.85, and ≥ 0.86; men: ≤ 0.95, 0.96–0.99, and ≥ 1.00, respectively) [38].
2.4. Statistical Analysis
Categorical variables are described using absolute values and proportions, and continuous variables using central tendency and dispersion measures. For comparisons between participants with and without T2DM, Chi‐square (X2) and independent t‐tests were used for categorical and continuous variables, respectively.
The agreement between the definition of BP phenotypes using different combinations of methods was assessed by computing the overall agreement and corresponding Cohen's kappa (κ). Sankey diagrams were used to illustrate the flow of participants between BP phenotypes defined using OBPM + ABPM and those defined with OBPM + u‐AOBP or OBPM + HBPM.
All the statistical tests were performed at two‐tailed 5% level of significance.
Data analysis was performed using IBM SPSS Statistics for Windows, Version 29.0 and the Sankey diagrams were generated using the SankeyMATIC builder [39].
3. Results
The characteristics of the participants are described in Table 1. In both groups 57% were female, with a mean age (standard deviation–SD) of 51.2 (10.4) years for participants with T2DM and 50.1 (10.0) years participants without T2DM. Among participants with T2DM, 66% had completed primary school or less, and only 7% had completed high school or university, while 29% of participants without T2DM had primary education or less, and 42% had completed high school or university (p < 0.001). Nearly one‐third of both groups had office SBP ≥ 140 mmHg and DBP ≥ 90 mmHg. More than two‐thirds in both groups were overweight or obese, while the WHR indicated a high cardiometabolic risk more frequently in T2DM patients (53.6% vs. 32.5%, p < 0.001).
TABLE 1.
Description of participants with and without type 2 diabetes mellitus (T2DM).
| Variable | Participants with T2DM | Participants without T2DM | p‐value b |
|---|---|---|---|
| (n = 151) a | (n = 151) a | ||
| Sex (Female) | 86 (57.0) | 86 (57.0) | > 0.999 |
| Age (years) c | |||
| ≤ 40 | 19 (12.6) | 21 (13.9) | 0.620 |
| 41–50 | 52 (34.4) | 61 (40.4) | |
| 51–60 | 50 (33.1) | 41 (27.2) | |
| ≥ 61 | 30 (19.9) | 28 (18.5) | |
| Education level | |||
| Up to primary education | 100 (66.2) | 43 (28.5) | < 0.001 |
| Secondary school completed | 40 (26.5) | 44 (29.1) | |
| University/post graduate degree | 11 (7.3) | 64 (42.4) | |
| Place of residence (Maputo City) | 63 (41.7) | 80 (53.0) | 0.050 |
| HIV infection (positive) | 25 (16.6) | 16 (10.6) | 0.131 |
| Office systolic BP | |||
| < 120 mmHg | 44 (29.1) | 47 (31.1) | 0.930 |
| 120–139 mmHg | 57 (37.7) | 55 (36.4) | |
| ≥ 140 mmHg | 50 (33.1) | 49 (32.5) | |
| Office diastolic BP | |||
| < 80 mmHg | 48 (31.8) | 54 (35.8) | 0.374 |
| 80–89 mmHg | 49 (32.5) | 38 (25.2) | |
| ≥ 90 mmHg | 54 (35.8) | 59 (39.1) | |
| Body mass index (Kg/m2) d | |||
| Underweight/normal weight | 41 (27.2) | 40 (26.5) | 0.970 |
| Overweight | 49 (32.5) | 51 (33.8) | |
| Obesity | 61 (40.4) | 60 (39.7) | |
| Waist‐hip ratio e | |||
| Low risk | 45 (29.8) | 75 (49.7) | < 0.001 |
| Moderate risk | 25 (16.6) | 27 (17.9) | |
| High risk | 81 (53.6) | 49 (32.5) | |
Data are presented as n (%), unless otherwise specified.
p‐values for categorical variables comparing participants with and without type 2 diabetes mellitus (T2DM) calculated using the Pearson's chi‐square test.
Mean age (Standard Deviation—SD)–T2DM: 51.2 (10.4); control: 50.1 (10.0); p = 0.171 (Student's t‐test).
Mean BMI (SD)–T2DM: 29.5 (7.1); control: 28.9 (5.6); p = 0.215 (Student's t‐test).
Mean WHR (SD)–T2DM: 0.92 (0.07); control: 0.88 (0.08; p < 0.001) (Student's t‐test). Cardiometabolic risk was classified according to sex‐specific WHR cut‐offs: low risk (women: ≤ 0.8; men: ≤ 0.95), moderate risk (women: 0.81–0.85; men: 0.96–0.99), high risk (women: ≥ 0.86; men: ≥ 1.0).
Serum sodium concentrations (mean [SD]) were significantly lower in participants with T2DM (136.51 [3.84] mmol/L vs. 138.74 [2.65] mmol/L; p < 0.001), whereas potassium levels were slightly higher (4.50 [0.56] mmol/L vs. 4.37 [0.40] mmol/L; p = 0.012). Participants with T2DM also had higher triglyceride (1.45 [0.94] mmol/L vs. 1.03 [0.58] mmol/L; p < 0.001) and LDL cholesterol levels (3.34 [1.17] mmol/L vs. 2.93 [0.93] mmol/L; p = 0.003) (Table S1).
3.1. Blood Pressure
BP measurements obtained by OBPM, u‐AOBP, daytime ambulatory ABPM, and HBPM were not significantly different between participants with and without T2DM. In contrast, nighttime ambulatory SBP and DBP were significantly higher in T2DM participants. Additionally, 24 h ambulatory SBP was significantly higher in participants with T2DM (128.5 vs. 124.4 mmHg; p = 0.045), while 24 h DBP did not differ between groups (p = 0.174) (Table S2).
3.2. Blood Pressure Phenotypes
The prevalence of BP phenotypes was significantly different between participants with and without T2DM (p = 0.005) only when the combination OBPM + ABPM was used; T2DM participants were less often classified as normotensive (32.5% vs 48.3%) but more likely to be classified as having masked hypertension (23.2% vs. 9.9%). Under OBPM + u‐AOBP combination, normotension predominated in both groups (53.6% in participants with T2DM vs. 57.6% in participants without T2DM), whereas masked hypertension remained rare (2.0% vs. 0.7%) (Table 2).
TABLE 2.
The BP phenotypes among participants with and without type 2 diabetes mellitus (T2DM), using different measurements methods combined with office blood pressure measurement (OBPM).
| Method of measurement | Participants with T2DM | Participants without T2DM | p‐value b |
|---|---|---|---|
| (n = 151) a | (n = 151) a | ||
| OBPM‡ | |||
| Normotension | 84 (55.6) | 88 (58.3) | 0.642 |
| Hypertension | 67 (44.4) | 63 (41.7) | |
| OBPM + u‐AOBP d | |||
| Normotension | 81 (53.6) | 87 (57.6) | 0.506 |
| White‐coat hypertension | 23 (15.2) | 27 (17.9) | |
| Masked hypertension | 3 (2.0) | 1 (0.7) | |
| Sustained hypertension | 44 (29.1) | 36 (23.8) | |
| OBPM + ABPM e | |||
| Normotension | 49 (32.5) | 73 (48.3) | 0.005 |
| White‐coat hypertension | 13 (8.6) | 11 (7.3) | |
| Masked Hypertension | 35 (23.2) | 15 (9.9) | |
| Sustained hypertension | 54 (35.8) | 52 (34.4) | |
| OBPM + HBPM f , g | |||
| Normotension | 66 (44.3) | 80 (53.3) | 0.149 |
| White‐coat hypertension | 16 (10.7) | 14 (9.3) | |
| Masked Hypertension | 18 (12.1) | 8 (5.3) | |
| Sustained hypertension | 49 (32.9) | 48 (32.0) | |
Data are presented as n (%), unless otherwise specified.
p‐values for the comparison of the distribution of BP phenotypes between participants with and without type 2 diabetes mellitus (T2DM) within each participant group calculated using the Pearson's chi‐square test.
OBPM: Office blood pressure measurement.
u‐AOBP: Unattended automated office blood pressure measurement.
ABPM: Ambulatory blood pressure monitoring.
HBPM: Home blood pressure monitoring.
Data available for 299 participants (T2DM: 149; control: 150).
In an additional comparison, BP phenotypes were also classified using u‐AOBP in combination with ABPM (u‐AOBP, + ABPM). Compared with the OBPM + ABPM classification, this resulted in a modest redistribution of phenotypes, with lower proportions of normotension (overall: 36.4% vs. 40.4%; participants with T2DM: 30.5% vs. 32.5%; participants without T2DM: 42.4% vs. 48.3%) and masked hypertension (overall: 13.6% vs. 16.6%; participants with T2DM: 17.9% vs. 23.2%; participants without T2DM: 9.3% vs. 9.9%), and higher proportions of white‐coat hypertension (overall: 11.9% vs. 7.9%; participants with T2DM: 10.6% vs. 8.6%; participants without T2DM: 13.2% vs. 7.3%) and sustained hypertension (overall: 38.1% vs. 35.1%; participants with T2DM: 41.1% vs. 35.8%; participants without T2DM: 35.1% vs. 34.4%).
According to participants' self‐reported history of hypertension, half of those with T2DM who reported never having been told by a health professional that they had raised BP were classified as hypertensive according to the OBPM + ABPM assessment (50/100, 50.0%), including 24 (24.0%) with masked hypertension and 26 (26.0%) with sustained hypertension. Among participants without T2DM, 31 of 94 (33.0%) without a previous history of hypertension fulfilled the criteria for hypertension, including 10 (10.6%) with masked hypertension and 21 (22.3%) with sustained hypertension. In both groups, the distribution of BP phenotypes differed significantly according to self‐reported history of hypertension (Table 3).
TABLE 3.
The BP phenotypes based on office blood pressure measurement (OBPM) and ambulatory blood pressure monitoring (ABPM), according to self‐reported previous diagnosis of hypertension.
| Variable | Participants with T2DM a | Participants without T2DM | ||
|---|---|---|---|---|
| (n = 151) | (n = 151) | |||
| Previous diagnosis of hypertension b | No | Yes | No | Yes |
| (n = 100) | (n = 51) | (n = 94) | (n = 57) | |
| OBPM + ABPM | ||||
| Normotension | 42 (42.0) | 6 (11.8) | 56 (59.6) | 14 (24.6) |
| White‐coat hypertension | 8 (8.0) | 6 (11.8) | 7 (7.4) | 7 (12.3) |
| Masked hypertension | 24 (24.0) | 10 (19.6) | 10 (10.6) | 3 (5.3) |
| Sustained hypertension | 26 (26.0) | 29 (56.9) | 21 (22.3) | 33 (57.9) |
| p‐value c | < 0.001 | < 0.001 | ||
Data are presented as n (%).
T2DM–Type 2 diabetes mellitus.
Previous diagnosis of hypertension was self‐reported.
p‐value for the comparison of the distribution of BP phenotypes according to self‐reported previous diagnosis of hypertension within each participant group calculated using the Pearson's chi‐square test.
3.3. OBPM Versus OBPM + u‐AOBP, OBPM + ABPM or OBPM + HBPM
Figure 1 shows the misclassification of BP phenotypes when using OBPM alone. In the Sankey diagrams, the width of each flow represents the number of participants reclassified across phenotypes.
FIGURE 1.
Misclassification of BP phenotypes when using office BP measurement (OBPM) alone, in relation to OBPM + unattended automated office BP (u‐AOBP), panels A and B; OBPM + ambulatory BP monitoring (ABPM), panels C and D, or OBPM + home BP monitoring (HBPM), panels E* and F†, among participants with and without type 2 diabetes mellitus (T2DM). * in panel E the overall number of participants is lower than 151 due to missing HBPM data for 2 participants with T2DM. † in panel F the overall number of participants is lower than 151 due to missing HBPM data for 1 participant without T2DM.
All combined methods revealed substantial reclassification, mainly from OBPM‐defined normotension to masked hypertension, particularly when ABPM was used. u‐AOBP and HBPM were useful to diagnose white‐coat hypertension, whereas HBPM identified some cases of masked hypertension, particularly among participants with T2DM, although with lower sensitivity than ABPM. Smaller transitions from hypertension to white‐coat hypertension were seen, particularly with u‐AOBP and ABPM. Overall, misclassification of the phenotypes was more pronounced in participants with T2DM.
3.4. Concordance between OBPM + ABPM and OBPM + u‐AOBP or OBPM + HBPM
As depicted in Figure 2, concordance with OBPM + ABPM was higher for OBPM + HBPM [T2DM: 73.2%; κ (95%CI): 0.62 (0.52–0.71); control: 81.3%; κ (95%CI): 0.70 (0.60–0.80)] than for OBPM + u‐AOBP (T2DM: 64.2%; κ (95%CI): 0.51 (0.40–0.61)].
FIGURE 2.
Concordance between BP phenotypes defined using office blood pressure measurement (OBPM) + ambulatory blood pressure monitoring (ABPM) and using OBPM + unattended automated office blood pressure (u‐AOBP), panels A and B; or OBPM + home blood pressure monitoring (HBPM), panels C and D, among participants with and without type 2 diabetes mellitus (T2DM).
Agreement was systematically lower among participants with T2DM for both comparisons. The diagrams also show that the largest discrepancies between OBPM + ABPM and the alternative combination of methods involved reclassification from masked hypertension to normotension, particularly in individuals with T2DM and more accentuated with u‐AOBP. Additionally, participants classified as having sustained hypertension by OBPM + ABPM were sometimes reassigned to white‐coat hypertension when using OBPM + u‐AOBP or OBPM + HBPM, although such flows were less frequent.
3.5. Dipping Pattern of Systolic BP
Figure 3 shows the distribution of SBP dipping patterns using ABPM. Overall, the distribution of dipping patterns differed significantly between participants with and without T2DM (p < 0.001). Among participants with T2DM, the prevalence of normal dipping was substantially lower than in participants without T2DM (27.2% vs. 53.3%). Conversely, non‐dipping and reverse‐dipping patterns were substantially more frequent in the T2DM group (56.3% and 14.6%, respectively) than among participants without T2DM (40.7% and 2.7% respectively). The proportion of extreme dippers was the lowest in both groups.
FIGURE 3.
Dipping pattern of systolic BP in participants with and without type 2 diabetes mellitus (T2DM).
4. Discussion
Our results show that, when using OBPM combined with ABPM to define BP phenotypes, just over one‐third of both participants in both groups had sustained hypertension, but the prevalence of masked hypertension was more than two‐fold higher among those with T2DM. This likely reflects the higher prevalence of a non‐dipping and reverse‐dipping patterns among participants with T2DM. Consistent with previous reports, while u‐AOBP reduces the white‐coat effect associated to OBPM, it may not fully capture the dynamic BP variability revealed by ABPM, particularly nocturnal patterns [40]. In fact, when complementing OBPM with u‐AOBP, sustained hypertension was often misclassified as white‐coat hypertension and masked hypertension as normotension, when compared with the gold standard (OBPM + ABPM). In contrast, the concordance between OBPM + HBPM tended to be better than when using OBPM + u‐AOBP, both for white‐coat and masked hypertension. However, when compared with OBPM + ABPM, its capacity to diagnose white‐coat is much better than the one to diagnose masked hypertension, probably reflecting the fact that HBPM does not evaluate night BP, and therefore its ability to diagnose masked hypertension is low.
A meta‐analyses reported a pooled prevalence of masked hypertension of 11% (95% CI: 4.7–19.3) across more than 7000 adults from several Sub‐Saharan African countries, with estimates ranging from 1.2% to 33.2% [41]. Our findings in individuals without T2DM are consistent with this range.
In contrast, the prevalence of masked hypertension among patients with T2DM has been more consistent across studies and similar to our findings. Reported estimates include 26.5% in a study of 266 participants and 32.5% in a more recent study of 77 untreated patients from Nigeria [22, 42].
An important observation in our study is that participants with and without T2DM, matched by age and sex, also showed similar BMI and mean BP values across OBPM, u‐AOBP, ABPM, and HBPM. Nevertheless, participants with T2DM were more frequently classified as being at high cardiometabolic risk based on WHR cut‐offs than participants without T2DM; these findings show an increase in central adiposity, despite similar overall adiposity, as assessed by BMI.
Although obesity and larger arm circumference may affect BP measurement when inappropriate cuff sizes are used, BP measurements in the present study were performed using cuffs appropriate for participants’ upper‐arm circumference: a cuff suitable for arm circumferences of 22–42 cm was used for OBPM and HBPM, whereas cuff size was selected according to arm circumference for u‐AOBP and ABPM. This is particularly relevant because Mourad et al. demonstrated that inappropriate cuff sizing during HBPM can substantially influence the prevalence of masked hypertension [43]. Furthermore, BMI distributions were comparable between participants with and without T2DM, making systematic differences in cuff size requirements an unlikely explanation for the higher prevalence of masked hypertension observed in the T2DM group. The main difference was the higher nighttime and 24 h SBP in participants with T2DM. This highlights the need to understand the differences in the phenotype classification when using different methods, given the importance of an accurate identification of the BP phenotypes for the definition of the therapeutic attitudes.
u‐AOBP, as used in the SPRINT trial [44], could be a simple and useful method to diagnose white‐coat hypertension, by measuring BP without the stressful presence of a health care worker. It is generally accepted that mean u‐AOBP BP is 5 to 10 mmHg lower than the mean OBPM and that this difference can increase to 10 to 20 mmHg when this last one is evaluated by an auscultatory method [8]. We have used an automated BP machine and the mean of the last two measurements in our study and found no statistically significant difference between the mean of u‐AOBP and OBPM. In the present investigation, u‐AOBP was able to identify about 15% of people with white‐coat hypertension, among the participants classified as hypertensive when using only OBPM. However, using OBPM + ABPM, almost half of these individuals were classified as hypertensive. Because nocturnal BP elevation is an important contributor to masked hypertension, u‐AOBP was not useful to detect masked hypertension both in individual with and without T2DM.
When u‐AOBP replaced the conventional OBPM as the office component of BP phenotype classification, the distribution of BP phenotypes changed modestly, with fewer participants classified as having masked hypertension and more classified as having sustained hypertension. Although this pattern differed from that reported by Bertram et al., important methodological differences should be considered, including differences in the study population, the use of 24 h rather than daytime ABPM for phenotype classification, and the objectives of the analyses [45].
HBPM, by capturing BP outside the clinical setting, improves detection of both white‐coat and masked hypertension when combined with OBPM, as shown in our study, especially among individuals with T2DM. Nevertheless, OBPM + HBPM still misclassified the BP phenotype of nearly one‐quarter of the individuals, as defined by OBPM + ABPM, especially those with masked hypertension, who are misclassified as normotensive.
A previous study comparing awake ABPM values with HBPM reported high concordance between HBPM and ABPM for both masked and white‐coat hypertension, with only 8.2% of discordant cases [46]. A more recent study conducted in a sample of 333 American adults not on antihypertensive therapy, showed that, using the 24 h ABPM values, the proportion of masked hypertension cases identified more than doubled compared with those identified using HBPM (25.8% vs. 11.1%) [47].
Our findings suggest that, although u‐AOBP and HBPM offer practical alternatives to ABPM, particularly in low‐resource settings, neither method provides an accurate characterization of BP phenotypes. HBPM showed better agreement with ABPM than u‐AOBP for the identification of both white‐coat and masked hypertension, but still failed to detect a substantial proportion of masked hypertension, particularly cases related to nocturnal BP elevation. u‐AOBP may help reduce the white‐coat effect during office evaluation and improve the identification of white‐coat hypertension; however, it does not adequately capture BP variability outside the clinical setting and therefore performs poorly in identifying masked hypertension. From both a clinical and public health perspective, our findings support prioritizing ABPM for individuals with T2DM because of its superior ability to detect masked hypertension and abnormal nocturnal BP patterns. When ABPM is unavailable, HBPM represents the most practical alternative for out‐of‐office BP assessment, although clinicians should recognize that normal HBPM or u‐AOBP values do not reliably exclude masked hypertension.
Notably, half of the participants with recently diagnosed T2DM who reported no previous history of hypertension were found to have sustained or masked hypertension. This finding highlights the potential for substantial under‐recognition of hypertension in this population and reinforces the importance of out‐of‐office BP assessment, particularly ABPM, to improve detection of clinically relevant BP phenotypes.
We also observed a higher prevalence of non‐dipping and reverse‐dipping patterns among participants with T2DM and reduced prevalence of normal dipping patterns, indicating a significant disturbance in nocturnal BP regulation. These alterations likely reflect autonomic dysfunction and vascular impairment associated with hyperglycaemia and insulin resistance. Consistent with previous findings, up to two‐thirds of individuals with T2DM exhibit blunted nocturnal declines in BP, which are linked to elevated cardiovascular risk [23, 48, 49]. Another possible explanation for the high prevalence of non‐dipping pattern could also be related to sleep apnea that is known to be common among obese people [50]. However, sleep apnea was not systematically assessed in the present study, and therefore its potential contribution to the observed nocturnal BP abnormalities could not be evaluated. This again underscores the clinical importance of using the 24 h ABPM monitoring in T2DM, as conventional office measurements may fail to capture this high‐risk phenotype and HBPM also fails to detect it in most cases. What is interesting to note in our data is the large proportion of individuals with recently diagnosed T2DM that present a non‐dipper or reverse dipper pattern, which can be interpreted in two ways: either it is present since the very beginning of the T2DM disease or, probably more plausible here, it is the consequence of a late diagnosis after a long asymptomatic period.
Strengths of this study include the simultaneous assessment of multiple BP measurement methods in both participants with and without T2DM, within the same urban African setting. Moreover, this study provided evidence the use of BP phenotypes enhances clinical relevance, which aligns with recent narrative reviews emphasizing gaps in hypertension detection, treatment and control in African populations [51].
This study has some limitations. First, although participants had a recent clinical diagnosis of T2DM, the true duration of disease before diagnosis could not be established. Consequently, some participants may already have had diabetes‐related microvascular or macrovascular complications at study entry. These complications were not systematically assessed and therefore could not be accounted for in the analyses. Second, the study was conducted in an urban Mozambican context, which may limit the generalizability of our findings to rural populations or other African settings.
Furthermore, although all participants with diabetes fulfilled the diagnostic criteria for T2DM defined for the study, we cannot exclude the presence of atypical diabetes subtypes described in African populations. Ketosis‐prone diabetes, the most widely recognized atypical subtype, is typically characterized by marked hyperglycaemia and presentation with ketosis or diabetic ketoacidosis. None of the participants presented these clinical features at recruitment; although this does not completely exclude atypical diabetes, it makes this particular subtype less likely in our study population [52, 53].
Overall, our findings highlight the clinical and public health importance of accurately characterizing BP phenotypes among individuals with T2DM in Sub‐Saharan African settings. The high prevalence of masked hypertension and abnormal nocturnal BP patterns observed in participants with T2DM suggests that reliance on OBPM alone may substantially underestimate cardiovascular risk in this population. Expanding access to ABPM could therefore play a key role in improving hypertension detection and guiding more appropriate therapeutic decisions. In settings where ABPM remains limited, pragmatic strategies combining u‐AOBP measurement with HBPM may represent a feasible interim approach, although clinicians should remain aware of the risk of misclassification, particularly for masked hypertension. From a broader perspective, these findings underscore the need for health systems in low‐resource contexts to strengthen hypertension surveillance and integrate phenotype‐based assessment into diabetes care pathways. Future studies should evaluate cost‐effective strategies to expand access to out‐of‐office BP monitoring and determine whether earlier identification of masked hypertension and nocturnal BP abnormalities can translate into improved cardiovascular outcomes among African populations with T2DM.
5. Conclusion
This study demonstrated that BP phenotypes are differentially distributed between individuals with recently diagnosed T2DM and those without T2DM, with sustained and masked hypertension being particularly prevalent among the T2DM group. ABPM remains the most reliable approach for capturing the full spectrum of BP phenotypes, especially masked hypertension, which holds significant prognostic relevance for cardiovascular and diabetes‐related outcomes.
Participants with T2DM also exhibited fewer normal dippers and more non‐dipping and reverse‐dipping patterns, indicating impaired nocturnal BP regulation and increased cardiovascular risk. Although u‐AOBP and HBPM provided valuable complementary information, both methods showed limitations in identifying white‐coat and masked hypertension, underscoring the potential for diagnostic misclassification when ABPM is not utilized.
These findings emphasize the importance of incorporating ABPM into diagnostic and management pathways for individuals with T2DM, while recognizing the practical utility of u‐AOBP and HBPM in resource‐limited settings. Expanding access to accurate BP assessment tools is crucial to enable early detection, guide timely intervention, and reduce the cardiovascular burden in this high‐risk population.
Funding
This work was supported by the Calouste Gulbenkian Foundation through a scholarship awarded to V.G.
Ethics Statement
This study was approved by the Comité Nacional de Bioética para Saúde (reference 313/CNBS/21).
Consent
All participants provided written informed consent prior to participation.
Conflicts of Interest
The authors declare no conflicts of interest.
Permission to Reproduce Material
No previously published material requiring permission for reproduction was used in this manuscript.
Clinical Trial Registration
The authors have nothing to report.
Supporting information
Supporting Information: Supplemental material 1: Biochemical parameters in participants with and without type 2 diabetes mellitus.
Supporting Information: Supplemental material 2: Blood pressure measurements in participants with and without type 2 diabetes mellitus.
Acknowledgments
We thank the participating health facilities for their support and collaboration in participant identification and recruitment. We also acknowledge the physicians and technical staff for their essential contributions to the implementation of the study procedures.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Mousavi S. S., Reyna M. A., Clifford G. D., and Sameni R., “A Survey on Blood Pressure Measurement Technologies: Addressing Potential Sources of Bias,” Sensors 24, no. 6 (2024): 1730. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Whelton P. K., Carey R. M., Aronow W. S., et al., “2017 ACC/AHA/ AAPA/ ABC/ ACPM/ AGS/ APhA/ ASH/ ASPC/ NMA/ PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines,” Hypertension 71, no. 6 (2018): e13–e115. [DOI] [PubMed] [Google Scholar]
- 3. Stergiou G. S., Asayama K., Thijs L., et al., “Prognosis of White‐Coat and Masked Hypertension: International Database of Home Blood Pressure in Relation to Cardiovascular Outcome,” Hypertension 63, no. 4 (2014): 675–682. [DOI] [PubMed] [Google Scholar]
- 4. Tewari J., Qidwai K. A., Roy S., et al., “Different Phenotypes of Hypertension and Associated Cardiovascular and all‐cause Mortality: A Systematic Review and Meta‐analysis,” The Egyptian Heart Journal 76, no. 1 (2024): 162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Chia J., Bhatia K. S., Mihailidou A. S., and Kanagaratnam L. B., “The Role of Ambulatory Blood Pressure Monitoring in Current Clinical Practice,” Heart, Lung and Circulation 31, no. 10 (2022): 1333–1340. [DOI] [PubMed] [Google Scholar]
- 6. Siddique S., Hameed Khan A., Shahab H., et al., “Office Blood Pressure Measurement: A Comprehensive Review,” Journal of Clinical Hypertension 23, no. 3 (2021): 440–449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Vischer A. S. and Burkard T., “Principles of Blood Pressure Measurement—Current Techniques, Office vs Ambulatory Blood Pressure Measurement,” Advances in Experimental Medicine and Biology 956 (2017): 85–96. [DOI] [PubMed] [Google Scholar]
- 8. Mancia G., Brunström M., Burnier M., et al., “2023 ESH Guidelines for the management of arterial hypertension The Task Force for the management of arterial hypertension of the European Society of Hypertension Endorsed by the European Renal Association (ERA) and the International Society of Hypertension (ISH),” Journal of Hypertension 41, no. 12 (2023): 1874–2071. [DOI] [PubMed] [Google Scholar]
- 9. Salvetti M., Paini A., Aggiusti C., et al., “Unattended versus Attended Blood Pressure Measurement,” Hypertension 73, no. 3 (2019): 736–742. [DOI] [PubMed] [Google Scholar]
- 10. Narkiewicz K., “Unattended Blood Pressure Measurement: Is It Really Superior to Traditional Clinic Readings?,” Polskie Archiwum Medycyny Wewnętrznej 134 (2024): 1–2. [DOI] [PubMed] [Google Scholar]
- 11. Jones D. W., Ferdinand K. C., Taler S. J., et al., “2025 AHA/ACC/AANP/AAPA/ABC/ACCP/ACPM/AGS/AMA/ASPC/NMA/PCNA/SGIM Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults,” Journal of the American College of Cardiology 86, no. 18 (2025): 1567–1678. [DOI] [PubMed] [Google Scholar]
- 12. Williams B., Mancia G., Spiering W., et al., “2018 ESC/ESH Guidelines for the Management of Arterial Hypertension,” European Heart Journal 39, no. 33 (2018): 3021–3104. [DOI] [PubMed] [Google Scholar]
- 13. Sun H., Saeedi P., Karuranga S., et al., “IDF Diabetes Atlas: Global, Regional and Country‐Level Diabetes Prevalence Estimates for 2021 and Projections for 2045,” Diabetes Research and Clinical Practice 183 (2022): 109119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Magliano D. J. and Boyko E. J., “IDF DIABETES ATLAS,” in IDF DIABETES ATLAS (International Diabetes Federation, 2021). [Google Scholar]
- 15. Silva‐Matos C., Gomes A., Azevedo A., Damasceno A., Prista A., and Lunet N., “Diabetes in Mozambique: Prevalence, Management and Healthcare Challenges,” Diabetes & Metabolism 37, no. 3 (2011): 237–244. [DOI] [PubMed] [Google Scholar]
- 16. Damasceno A., Cubula B., and Lunet N., “Relatório final: Avaliação dos factores de risco das Doenças Não Transmissíveis na população moçambicana – STEPS 1,2,3 Moçambique 2014/2015.” Maputo. (2017).
- 17. Ohishi M., “Hypertension With Diabetes Mellitus: Physiology and Pathology,” Hypertension Research 41, no. 6 (2018): 389–393. [DOI] [PubMed] [Google Scholar]
- 18. Jia G. and Sowers J. R., “Hypertension in Diabetes: An Update of Basic Mechanisms and Clinical Disease,” Hypertension 78, no. 5 (2021): 1197–1205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Colussi G., Porto A. D. A., and Cavarape A., “Hypertension and Type 2 Diabetes: Lights and Shadows About Causality,” Journal of Human Hypertension 34, no. 2 (2020): 91–93. [DOI] [PubMed] [Google Scholar]
- 20. World Health Organization , Guidelines for the Prevention, Management and Care of Diabetes Mellitus, Khatib O. M., Editor. (EMRO Technical Publications Series; 32 2006): 82. [Google Scholar]
- 21. Hadjkacem F., Triki F., Frikha H., et al., “[Masked Arterial Hypertension in Patients With type2 Diabetes Mellitus: Prevalence, Associated Factors and Cardiovascular Impact],” Annales De Cardiologie Et D angeiologie (Paris) 71, no. 3 (2022): 136–140. [DOI] [PubMed] [Google Scholar]
- 22. Zhao H., Zeng F., Wang X., and Wang L., “Prevalence, Risk Factors, and Prognostic Significance of Masked Hypertension in Diabetic Patients,” Medicine 96, no. 43 (2017): e8363. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Călin P., Viorel M., Luchiana P., Mihaela C., and Lavinia P., “Masked Nocturnal Hypertension as a Result of High Prevalence of Non‐Dippers Among Apparently Well‐controlled Hypertensive Patients With Type 2 Diabetes Mellitus: Data From a Prospective Study,” Diabetology & Metabolic Syndrome 14, no. 1 (2022): 130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Brown J. B., Ramaiya K., Besancon S., et al., “Use of Medical Services and Medicines Attributable to Diabetes in Sub‐Saharan Africa,” PLoS ONE 9, no. 9 (2014): e106716. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Beckman J. A. and Creager M. A., “Vascular Complications of Diabetes,” Circulation Research 118, no. 11 (2016): 1771–1785. [DOI] [PubMed] [Google Scholar]
- 26. Lago R. M., Singh P. P., and Nesto R. W., “Diabetes and Hypertension,” Nature Clinical Practice Endocrinology & Metabolism 3, no. 10 (2007): 667–668. [DOI] [PubMed] [Google Scholar]
- 27. Silva A., Cruz R. C., Albuquerque N. L. S., Silva V. M. D., and Araujo T. L., “Blood Pressure Variability in Individuals With Diabetes Mellitus: A Scoping Review,” Revista Brasileira De Enfermagem 75, no. 6 (2022): e20210804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Instituto Nacional de Estatística “IV Recenseamento geral da População e habitação 2017. Resultados definitivos—Moçambique,” in Instituto Nacional de Estatística, Direcção de Estatísticas Demográficas (Vitais e Sociais: Maputo, Moçamique, 2019), 214. [Google Scholar]
- 29. Cambaza E., “Mozambique: Country Profile,” Encyclopedia 3, no. 1 (2023): 143–167. [Google Scholar]
- 30. Cavoli C., Oviedo D., Yusuf Y., et al., Opportunities and Barriers to Transition Towards Sustainable Mobility in Sub‐Saharan African Cities: The Case of Maputo and Freetown Available at SSRN 4919723, 2024.
- 31. McCordic C. and Abrahamo E., “Family Structure and Severe Food Insecurity in Maputo and Matola, Mozambique,” Sustainability 11, no. 1 (2019): 267. [Google Scholar]
- 32. World Health Organization , Classification of Diabetes Mellitus 2019.
- 33. Alper B. S., Price A., van Zuuren E. J., et al., “Consistency of Recommendations for Evaluation and Management of Hypertension,” JAMA Network Open 2, no. 11 (2019): e1915975–e1915975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Parati G., Stergiou G. S., Asmar R., et al., “European Society of Hypertension Practice Guidelines for Home Blood Pressure Monitoring,” Journal of Human Hypertension 24, no. 12 (2010): 779–785. [DOI] [PubMed] [Google Scholar]
- 35. Lin Y. T., Lampa E., Fall T., Engström G., and Sundström J., “Blood Pressure Phenotypes Based on Ambulatory Monitoring in a General Middle‐Aged Population,” Blood Pressure 30, no. 4 (2021): 237–249. [DOI] [PubMed] [Google Scholar]
- 36. Tocci G., Citoni B., Nardoianni G., Figliuzzi I., and Volpe M., “Current Applications and Limitations of European Guidelines on Blood Pressure Measurement: Implications for Clinical Practice,” Internal and Emergency Medicine 17, no. 3 (2022): 645–654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. World Health Organization , Obesity: Preventing and Managing the Global Epidemic, World Health Organization Technical Report Series 894 (2000): 1–253. [PubMed] [Google Scholar]
- 38. World Health Organization , Waist circumference and waist‐hip ratio: Report of a WHO Expert Consultation, 8–11 2008., (World Health Organization, 2011). [Google Scholar]
- 39.Available from: Bogart S. SankeyMATIC. 2025; https://sankeymatic.com/.
- 40. Colantonio L. D., Anstey D. E., Carson A. P., et al., “Metabolic Syndrome and Masked Hypertension Among African Americans: The Jackson Heart Study,” Journal of Clinical Hypertension (Greenwich) 19, no. 6 (2017): 592–600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Noubiap J. J., Nansseu J. R., Nkeck J. R., Nyaga U. F., and Bigna J. J., “Prevalence of White Coat and Masked Hypertension in Africa: A Systematic Review and Meta‐Analysis,” Journal of clinical hypertension 20, no. 8 (2018): 1165–1172, (Greenwich, Conn.). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Oshaju H., Chori B. S., Mshelia‐Reng R., Anumah F. O., and Odili A. N., “Prevalence and Determinants of Masked Hypertension Among Patients With Type 2 Diabetes Mellitus,” Journal of Human Hypertension 39, no. 12 (2025): 867–873. [DOI] [PubMed] [Google Scholar]
- 43. Mourad J. J., Lopez‐Sublet M., Aoun‐Bahous S., et al., “Impact of Miscuffing During Home Blood Pressure Measurement on the Prevalence of Masked Hypertension,” American Journal of Hypertension 26, no. 10 (2013): 1205–1209. [DOI] [PubMed] [Google Scholar]
- 44. Kjeldsen S. E., Lund‐Johansen P., Nilsson P. M., and Mancia G., “Unattended Blood Pressure Measurements in the Systolic Blood Pressure Intervention Trial: Implications for Entry and Achieved Blood Pressure Values Compared With Other Trials,” Hypertension 67, no. 5 (2016): 808–812. [DOI] [PubMed] [Google Scholar]
- 45. Bertram S., Bauer F., Shadi R., et al., “Prevalence of Masked Hypertension in Attended versus Unattended Office Blood Pressure Measurement,” Journal of Clinical Hypertension 26, no. 6 (2024): 615–623. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Nasothimiou E. G., Tzamouranis D., Rarra V., Roussias L. G., and Stergiou G. S., “Diagnostic Accuracy of Home vs. ambulatory Blood Pressure Monitoring in Untreated and Treated Hypertension,” Hypertension Research 35, no. 7 (2012): 750–755. [DOI] [PubMed] [Google Scholar]
- 47. Anstey D. E., Muntner P., Bello N. A., et al., “Diagnosing Masked Hypertension Using Ambulatory Blood Pressure Monitoring, Home Blood Pressure Monitoring, or Both?,” Hypertension 72, no. 5 (2018): 1200–1207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Cuspidi C., Sala C., Tadic M., Gherbesi E., Grassi G., and Mancia G., “Nondipping Pattern and Carotid Atherosclerosis: A Systematic Review and Meta‐Analysis,” Journal of Hypertension 34, no. 3 (2016): 385–392. [DOI] [PubMed] [Google Scholar]
- 49. Matteucci E. and Giampietro O., “Ambulatory Blood Pressure Monitoring and Circadian Rhythm of Blood Pressure in Diabetes Mellitus,” European Medical Journal Diabetes 1 (2013): 38–43. [Google Scholar]
- 50. Kwon Y., Tzeng W. S., Seo J., et al., “Obstructive Sleep Apnea and Hypertension; Critical Overview,” Clinical Hypertension 30, no. 1 (2024): 19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Gafane‐Matemane L. F., Craig A., Kruger R., et al., “Hypertension in Sub‐Saharan Africa: The Current Profile, Recent Advances, Gaps, and Priorities,” Journal of Human Hypertension 39, no. 2 (2025): 95–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Katte J. C., Bavuma C., Wild S. H., et al., “Atypical Diabetes Subtypes in Black African Populations,” Diabetologia 69, no. 6 (2026): 1432–1443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Bavuma C., Sahabandu D., Musafiri S., Danquah I., McQuillan R., and Wild S., “Atypical Forms of Diabetes Mellitus in Africans and Other Non‐European Ethnic Populations in Low‐ and Middle‐Income Countries: A Systematic Literature Review,” Journal of Global Health 9, no. 2 (2019): 020401. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information: Supplemental material 1: Biochemical parameters in participants with and without type 2 diabetes mellitus.
Supporting Information: Supplemental material 2: Blood pressure measurements in participants with and without type 2 diabetes mellitus.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.