New Diagnostic Criteria for Overmedication in Systemic Arterial Hypertension

The study was conducted in 2018 at the Unicordis Cardiologic Research Unit in Fortaleza, Brazil. Patients who presented for therapeutic evaluation underwent ABPM to detect HTN. Patients aged 18 through 75 years who were taking medication to control HTN were enrolled in the study.

Participants were given a Dinamap ABPM device (GE Healthcare) to use for monitoring their BP at home. The ABPM device was programmed to measure BP every 15 minutes while participants were awake and every 20 minutes as they slept.

At the clinic visit, the technician explained the study procedures and recorded the participant's height, weight, and arm circumference (for cuff size). Arterial pressure in both arms was measured in standard fashion while the person was sitting (after resting for 5 min) or standing (if older adult). If the SBP difference between arms was <10 mm Hg, the technician placed the cuff on the nondominant arm 2 to 3 cm above the elbow pit, programmed the monitor according to each participant's stated awake and sleep periods, compared ABPM and sphygmomanometer measurements to ensure agreement, and performed 2 trial measurements.

Participants were asked to provide a list of their medications, the dosages, and time(s) of day when taken. They were also instructed to maintain their regular medication schedule and habitual activity level but to avoid physical exercise during the 24 hours before the start of monitoring. Participants were shown how to deflate the ABPM cuff manually and how to take manual measurements in case of symptoms. They were asked not to disconnect the monitor at any time; to keep the arm still and relaxed during measurements; to readjust the cuff during the day and reposition the monitor before sleep, as needed; and to record their meal times, consumption of stimulants (namely, alcohol, cigarettes, and coffee), medications taken (name, dose, and time), methods and times of locomotion, stressful events, symptoms (onset and duration), awake and sleep periods, and sleep quality.

The study protocol was approved by the Research Ethics Committee of Hospital Universitário Walter Cantídio, Federal University of Ceará, Brazil (protocol no. 2.489.575). All participants gave written informed consent.

Quantitative variables are expressed as mean ± SD and categorical variables as number and percentage. The groups were compared using Student t or Mann-Whitney U test, depending on the normality of the data distribution. Associations were analyzed using the Pearson c2 test (for quantitative variables) or the Fisher exact test (for categorical variables). All analyses were performed with R software, version 3.3.1. P <0.05 was considered statistically significant.

During the study period, of 2,352 patients who underwent ABPM for noninvasive cardiologic testing, 1,105 underwent testing for therapeutic evaluation. Of these, 567 had medically controlled HTN and were enrolled in the study.

We divided the study sample into groups according to the percentage of valid SBP measurements less than 100 mm Hg. Group A comprised 413 participants (72.8% of the study cohort) in whom less than 10% of the measurements were less than 100 mm Hg, signifying no or mild hypotension and no overmedication. Group B comprised 154 participants (27.2% of the total) in whom 10% or more of the measurements were less than 100 mm Hg, signifying hypotension and overmedication. Group C included 74 members (48.1%) of group B in whom ≥20% of measurements were less than 100 mm Hg, signifying hypotension and an “accentuated” level of overmedication. The remaining 80 members in group B were considered to be mildly overmedicated.

The 3 groups were statistically similar in sex distribution (Table I) and mean age (Table II), whereas groups A and C differed with regard to the mean number of antihypertensive drugs prescribed (Table III).

When mean SBP was used to define hypotension, the number of patients with hypotension was 35 (6.2% of the study cohort) based on the 24-hour mean; 65 (11.5%) based on the daytime mean; 24 (4.2%) based on the nighttime mean; and 76 (13.4%) based on any mean (Table IV).

In group B, 59 participants took only 1 antihypertensive drug (monotherapy), and 95 took a combination (65 patients took 2 drugs, 26 took 3 drugs, and 4 took 4 drugs). The predominant drug classes were an angiotensin II receptor blocker (ARB) in monotherapy, an ARB plus a thiazide in dual therapy, an ARB plus a thiazide plus a calcium channel blocker (CCB) in triple therapy, and an ARB plus a CCB plus a β-blocker plus thiazide in quadruple therapy (Table V).

Diagnosing overmedication-induced hypotension depends on the criteria used to define hypotension. In this study, we used the percentage of SBP measurements less than 100 mm Hg during a 24-hour period as the standard; patients with hypotension and who were overmedicated were those in whom ≥10% of measurements were less than 100 mm Hg (group B). Within group B, those for whom 20% or more of measurements were less than 100 mm Hg were considered to be overmedicated at an accentuated level and were analyzed separately (group C). In our sample of 567 patients with medication-controlled HTN, 27.2% had hypotension (group B), and almost half of those were overmedicated at an accentuated level (group C). In contrast, when mean SBP was used as a criterion for hypotension, the 24-hour mean identified only 6.2% of participants as hypotensive, the daytime mean identified 11.5%, and the nighttime mean identified 4.2%, closely matching the findings of an earlier study (6.8%, 12.2%, and 3.9%, respectively).⁵ 

As might be expected, the mean number of antihypertensive drugs prescribed was greater in group C than in group A (ie, those with mild or no hypotension); otherwise, the 3 groups were similar with regard to sex distribution and mean age. The mean number of antihypertensive drugs was also positively associated with hypotension in a study by Divisón-Garrote et al.⁴  Moreover, according to McCracken et al,⁹  not only the number of antihypertensive drugs but also the number of drugs prescribed for other diseases may be correlated with hypotension. Barochiner et al¹⁰  found that old age, diabetes, and coronary disease were associated with an increased prevalence of hypotension. We did not collect information on comorbidities in the present study.

To our surprise, a substantial proportion of patients who were prescribed monotherapy (38.3%) presented with arterial hypotension. One explanation is that some were originally misdiagnosed as having HTN, making it more likely for a single antihypertensive drug to cause hypotension. The drug class most frequently taken by patients receiving monotherapy was ARBs, followed by β-blockers, angiotensin-converting enzyme inhibitors, CCBs, and thiazides; ARBs were also taken most often in multidrug combinations. Baena Díez et al⁷  reported that arterial hypotension among patients older than 80 years was most prevalent in those prescribed dual therapy (46.3%), followed by monotherapy (33.6%) and triple therapy (22.1%). In that cohort, the most frequently prescribed type of drug was thiazide, followed by angiotensin-converting enzyme inhibitors, ARBs, CCBs, and β-blockers.⁷ 

We propose that the number or percentage of SBP measurements less than 100 mm Hg, and not mean SBP on ABPM, be used to diagnose overmedication in HTN treatment. To illustrate the point, a patient taking an antihypertensive medication for whom 49% of SBP measurements were 99 mm Hg and 51% were 101 mm Hg would be hypotensive (<100 mm Hg) almost half the time; nevertheless, according to the conventional criteria defining overmedication (mean SBP, <100 mm Hg), the patient would not be considered overmedicated. With use of our proposed criteria, such a patient would be considered overmedicated at an accentuated level.

Of note, we defined arterial hypotension as an SBP less than 100 mm Hg regardless of the time of day. We agree with Owens and O'Brien,⁶  who proposed that this cutoff value is most adequate for the awake state, whereas a lower cutoff (90 mm Hg) should be used for ABPM measurements during sleep. However, we disagree with their opinion that diastolic pressure (which is inaccurate at lower ranges) is a useful criterion for determining hypotension.

This study is not the first to investigate or discuss overmedication in patients with HTN.^4–11  However, to our knowledge, we are the first to propose a classification system for overmedication.

We could not ask participants to record events (eg, falls, presyncope or syncope, and orthostatism), so we could not report clinical outcomes. Accordingly, a more systematic study is needed to evaluate our proposed classification system.

We propose the following 2-part definition of over-medication in the treatment of systemic arterial HTN: (1) mild overmedication, when, over a 24-hour period, 10% or more but less than 20% of SBP measurements are less than 100 mm Hg in the awake state and are less than 90 mm Hg during sleep; and (2) accentuated overmedication, when, over a 24-hour period, 20% or more of SBP measurements are less than 100 mm Hg in the awake state and are less than 90 mm Hg during sleep. Clear, proven definitions like these can help clinicians make better therapeutic choices. Overmedication in HTN treatment should be avoided, because too low of an SBP is associated with fainting and falls.