Pharmacological Management of Orthostatic Hypotension
Pharmacological Management of Orthostatic Hypotension
Although postural hypotension is a common problem in elderly people with significant morbidity and mortality, there appears to be little high-quality data as to the best pharmacological management. We report a systematic review of the results from 13 blinded RCTs, which examined the effects of drug treatment for OH. Of the 708 reports screened, 97 full-text articles were examined, but only 13 fitted our entry criteria, eight involving long-term treatment and five short-term or single-dose studies. There was a general paucity of good quality trials with comparable data parameters, which precluded a good quality meta-analysis. There was a considerable difference in effect on postural BP fall not only between trials (despite using the same pharmacological agent at the same dosages) whether short- or long-term effects were studied, but perhaps more predictably between agents. In general, these trials did show treatment increased standing or HUT SBP levels, but there was limited evidence of a greater clinical benefit of any specific therapeutic regime.
Current guidelines recognise the limited availability of prospective randomised controlled trials. The European Federation of Neurological Societies 2006 guidelines recommend fludrocortisone as first-line treatment and then midodrine, on its own or combined with fludrocortisone, after non-pharmacological measures have been tried including education and physical measures. They also recommend DOPS (dihydroxyphenylserine) and octreotide for the treatment of OH, which are also included in the recommendations in the European Handbook of Neurological Management in 2011. The ESC 2009 guidelines similarly recommend non-pharmacological measures in the treatment of OH, including adequate hydration and salt intake, as first-line management followed by midodrine and fludrocortisone along with pyridostigmine. Other drugs such as octreotide were also proposed where hypotension may be as a result of postprandial haemodynamic changes or erythropoietin where anaemia was the underlying cause.
Potential confounding factors that may have had a significant influence on the effects of the different pharmacological treatments on orthostatic BP change, as well as in symptoms, will have been the variation in the aetiology of the OH and the differing mechanisms of actions of the various agents. For example, fludrocortisone (9-alpha fludrohydrocortisone acetate) acts not only by increasing plasma volume by its sodium retaining effects as a synthetic mineralcorticoid thus increasing cardiac output but also by potentially increasing sensitivity to sympathetic nerve stimulation resulting in an increase peripheral vascular resistance; this latter effect being independent of norepinenephrine release from the sympathetic nerve endings in response to HUT. Midodrine is a pro-drug and its active agent desglymidodrine is an alpha-1-adrenoceptor agonist, which also increases mean systemic arterial pressure by raising peripheral vascular resistance. Future trials should take into account the likely aetiology of OH to select the agent with the most appropriate pharmacological profile. This could be aided by classifying OH according to changes in total peripheral resistance and cardiac output to determine whether OH is resulting from arteriolar, venous or mixed dysfunction.
Whether any improvement in the systemic orthostatic BP fall with treatment is associated with symptomatic improvement is less well known. We considered carrying out a detailed systematic review, but only some of the RCTs attempted to examine the effect of pharmacological intervention on symptoms; however, there was no consistency between these studies in terms of methodology (e.g. questionnaire used). Midodrine has been reported to significantly reduce the incidence of a patient's inability to stand, and improve the global postural symptom score, with good concurrence between patient and investigator scores. In the case of fludrocortisone therapy, it has also been reported to result in subjective improvement although the studies were too small to draw firm conclusions.
Many clinical reviews highlight the benefit of drug therapy for OH, as well as the many non-pharmacological options. ESC guidelines of 2009 based the recommendation of the use of midodrine in the treatment of OH on three studies, which because of differences in parameters given made it difficult to meta-analyse. However, this systematic review, which included small and large studies of varying duration, highlighted that although there is evidence of some beneficial effect of treatment in reducing the postural BP fall, the benefits in terms of symptom relief were unclear especially as duration of therapy and underlying aetiology of OH differed considerably between studies. The data on the benefits of long-term pharmacological treatment of OH are limited both in terms of the effects on postural BP changes as well as symptom relief and should be weighed against potential side effects and adverse effects including cardiac failure, systolic hypertension and stroke.
There are limitations to this review. There was significant variability in the definition of OH between the studies, with some groups including participants with only symptoms (in recognition that the actual fall in systemic BP may be limited), whilst others requiring in fall in postural BP greater than the current ESC guidelines on Syncope or the Consensus Statement. This highlights the suspicion that it may not be the fall in systemic BP that causes symptoms, but the failure to maintain cerebral blood flow as a result of impaired auto-regulation resulting in a fall in cerebral perfusion to the drop in systemic BP that is the underlying problem. Thus, treatments that are used solely to increase systemic BP levels may be inappropriate for some patients. There was a large variability in end-point parameters in studies involving the drug treatment of OH, making a meta-analysis comparing differing drugs and their effectiveness in treating the postural fall in BP impossible. We limited our study to reports that were published in the English language only.
The studies included in this review were carried out in Western Europe and North American and the results should be interpreted with caution as they may not be generalizable across other ethnic groups. Furthermore, the studies included young and older adults where the pathophysiology mechanisms may differ, with arterial stiffness and baroreceptor function affecting blood pressure with increasing age. In addition, the heterogeneity of the participants in the studies is high, with varying underlying causes of OH being included even within the same study.
There is always a possibility of publication bias with only positive effects being emphasised in clinical trials to date. Disappointingly, there was a lack of reporting of the amount of change in postural BP levels and standard deviations (i.e. the difference a drug exerted on the actual change from supine or sitting to standing or tilt blood pressure). Most studies did not report the magnitude of effect of drugs on in terms of the improvement or reduction (if any) in postural BP drop and more importantly none made any detailed comment on patients symptoms or quality of life factors.
We suggest that future trials should study the improvement in symptoms and QOL measures rather than concentrate just on changes in BP measurements. Using the ESC definition of OH in future studies as well as publishing standard deviations for changes in postural BP will allow comparison across studies. Of the few studies that did comment on the improvement in symptoms of OH with therapy, only Schoffer et al. used COMPASS-OD (questions relating specifically to OH and part of the Mayo clinic autonomic Symptom Profile), which is correlated with part of the Composite Autonomic Scoring Scale (CASS) in their outcome assessment.
The strength of this review is that we used strict selection criteria based on quality of methodology as well as reporting and three authors independently reviewed trials to select those eligible. In the absence of firm clinical evidence of the effect of pharmacological intervention for this common condition with its associated higher morbidity and mortality, non-pharmacological management may remain as an important first step towards the management of this condition although the evidence of their effectiveness is equally lacking. To our knowledge, there is only one open randomised trial of non-pharmacological management of OH in older patients to date, which concluded no benefit of a 6-inch head-of-bed elevation on BP or symptoms to other non-pharmacological management. A useful comparison for future trials could include pharmacological and non-pharmacological measures. Participants with OH could be randomised to a drug where a crossover of non-pharmacological measures is unsuccessful in terms of symptomatic relief and quality of life rather than BP improvement alone.
Discussion
Although postural hypotension is a common problem in elderly people with significant morbidity and mortality, there appears to be little high-quality data as to the best pharmacological management. We report a systematic review of the results from 13 blinded RCTs, which examined the effects of drug treatment for OH. Of the 708 reports screened, 97 full-text articles were examined, but only 13 fitted our entry criteria, eight involving long-term treatment and five short-term or single-dose studies. There was a general paucity of good quality trials with comparable data parameters, which precluded a good quality meta-analysis. There was a considerable difference in effect on postural BP fall not only between trials (despite using the same pharmacological agent at the same dosages) whether short- or long-term effects were studied, but perhaps more predictably between agents. In general, these trials did show treatment increased standing or HUT SBP levels, but there was limited evidence of a greater clinical benefit of any specific therapeutic regime.
Current guidelines recognise the limited availability of prospective randomised controlled trials. The European Federation of Neurological Societies 2006 guidelines recommend fludrocortisone as first-line treatment and then midodrine, on its own or combined with fludrocortisone, after non-pharmacological measures have been tried including education and physical measures. They also recommend DOPS (dihydroxyphenylserine) and octreotide for the treatment of OH, which are also included in the recommendations in the European Handbook of Neurological Management in 2011. The ESC 2009 guidelines similarly recommend non-pharmacological measures in the treatment of OH, including adequate hydration and salt intake, as first-line management followed by midodrine and fludrocortisone along with pyridostigmine. Other drugs such as octreotide were also proposed where hypotension may be as a result of postprandial haemodynamic changes or erythropoietin where anaemia was the underlying cause.
Potential confounding factors that may have had a significant influence on the effects of the different pharmacological treatments on orthostatic BP change, as well as in symptoms, will have been the variation in the aetiology of the OH and the differing mechanisms of actions of the various agents. For example, fludrocortisone (9-alpha fludrohydrocortisone acetate) acts not only by increasing plasma volume by its sodium retaining effects as a synthetic mineralcorticoid thus increasing cardiac output but also by potentially increasing sensitivity to sympathetic nerve stimulation resulting in an increase peripheral vascular resistance; this latter effect being independent of norepinenephrine release from the sympathetic nerve endings in response to HUT. Midodrine is a pro-drug and its active agent desglymidodrine is an alpha-1-adrenoceptor agonist, which also increases mean systemic arterial pressure by raising peripheral vascular resistance. Future trials should take into account the likely aetiology of OH to select the agent with the most appropriate pharmacological profile. This could be aided by classifying OH according to changes in total peripheral resistance and cardiac output to determine whether OH is resulting from arteriolar, venous or mixed dysfunction.
Whether any improvement in the systemic orthostatic BP fall with treatment is associated with symptomatic improvement is less well known. We considered carrying out a detailed systematic review, but only some of the RCTs attempted to examine the effect of pharmacological intervention on symptoms; however, there was no consistency between these studies in terms of methodology (e.g. questionnaire used). Midodrine has been reported to significantly reduce the incidence of a patient's inability to stand, and improve the global postural symptom score, with good concurrence between patient and investigator scores. In the case of fludrocortisone therapy, it has also been reported to result in subjective improvement although the studies were too small to draw firm conclusions.
Many clinical reviews highlight the benefit of drug therapy for OH, as well as the many non-pharmacological options. ESC guidelines of 2009 based the recommendation of the use of midodrine in the treatment of OH on three studies, which because of differences in parameters given made it difficult to meta-analyse. However, this systematic review, which included small and large studies of varying duration, highlighted that although there is evidence of some beneficial effect of treatment in reducing the postural BP fall, the benefits in terms of symptom relief were unclear especially as duration of therapy and underlying aetiology of OH differed considerably between studies. The data on the benefits of long-term pharmacological treatment of OH are limited both in terms of the effects on postural BP changes as well as symptom relief and should be weighed against potential side effects and adverse effects including cardiac failure, systolic hypertension and stroke.
There are limitations to this review. There was significant variability in the definition of OH between the studies, with some groups including participants with only symptoms (in recognition that the actual fall in systemic BP may be limited), whilst others requiring in fall in postural BP greater than the current ESC guidelines on Syncope or the Consensus Statement. This highlights the suspicion that it may not be the fall in systemic BP that causes symptoms, but the failure to maintain cerebral blood flow as a result of impaired auto-regulation resulting in a fall in cerebral perfusion to the drop in systemic BP that is the underlying problem. Thus, treatments that are used solely to increase systemic BP levels may be inappropriate for some patients. There was a large variability in end-point parameters in studies involving the drug treatment of OH, making a meta-analysis comparing differing drugs and their effectiveness in treating the postural fall in BP impossible. We limited our study to reports that were published in the English language only.
The studies included in this review were carried out in Western Europe and North American and the results should be interpreted with caution as they may not be generalizable across other ethnic groups. Furthermore, the studies included young and older adults where the pathophysiology mechanisms may differ, with arterial stiffness and baroreceptor function affecting blood pressure with increasing age. In addition, the heterogeneity of the participants in the studies is high, with varying underlying causes of OH being included even within the same study.
There is always a possibility of publication bias with only positive effects being emphasised in clinical trials to date. Disappointingly, there was a lack of reporting of the amount of change in postural BP levels and standard deviations (i.e. the difference a drug exerted on the actual change from supine or sitting to standing or tilt blood pressure). Most studies did not report the magnitude of effect of drugs on in terms of the improvement or reduction (if any) in postural BP drop and more importantly none made any detailed comment on patients symptoms or quality of life factors.
We suggest that future trials should study the improvement in symptoms and QOL measures rather than concentrate just on changes in BP measurements. Using the ESC definition of OH in future studies as well as publishing standard deviations for changes in postural BP will allow comparison across studies. Of the few studies that did comment on the improvement in symptoms of OH with therapy, only Schoffer et al. used COMPASS-OD (questions relating specifically to OH and part of the Mayo clinic autonomic Symptom Profile), which is correlated with part of the Composite Autonomic Scoring Scale (CASS) in their outcome assessment.
The strength of this review is that we used strict selection criteria based on quality of methodology as well as reporting and three authors independently reviewed trials to select those eligible. In the absence of firm clinical evidence of the effect of pharmacological intervention for this common condition with its associated higher morbidity and mortality, non-pharmacological management may remain as an important first step towards the management of this condition although the evidence of their effectiveness is equally lacking. To our knowledge, there is only one open randomised trial of non-pharmacological management of OH in older patients to date, which concluded no benefit of a 6-inch head-of-bed elevation on BP or symptoms to other non-pharmacological management. A useful comparison for future trials could include pharmacological and non-pharmacological measures. Participants with OH could be randomised to a drug where a crossover of non-pharmacological measures is unsuccessful in terms of symptomatic relief and quality of life rather than BP improvement alone.