Epidemiology of pulmonary hypertension: new data from the Swiss registry

Background: since 1999 data from pulmonary hypertension (PH) patients from all PH centres in Switzerland were prospectively collected. We analyse the epidemiological aspects of these data. Methods: PH was defined as a mean pulmonary artery pressure of >25 mm Hg at rest or >30 mm Hg during exercise. Patients with pulmonary arterial hypertension (PAH), PH associated with lung diseases, PH due to chronic thrombotic and/or embolic disease (CTEPH)

Background: since 1999 data from pulmonary hypertension (PH) patients from all PH centres in Switzerland were prospectively collected. We analyse the epidemiological aspects of these data.
Methods: PH was defined as a mean pulmonary artery pressure of >25 mm Hg at rest or >30 mm Hg during exercise. Patients with pulmonary arterial hypertension (PAH), PH associated with lung diseases, PH due to chronic thrombotic and/or embolic disease (CTEPH), or PH due to miscellaneous disorders were registered. Data from adult patients included between January 1999 and December 2004 were analysed.
Discussion: PH is a disease affecting adults of all ages. The management of these patients in specialised centres guarantees a high quality of care. Analysis of the registry data could be an instrument for quality control and might help identify weak points in assessment and treatment of these patients.

Introduction 379
Financial support: The Swiss Society for Pulmonary Hypertension received financial support from Actelion Switzerland to run the registry. from PH patients from all PH centres and associated hospitals in Switzerland were prospectively collected. In this study, we analyse the epidemio-logical aspects of these data and focus on basic clinical characteristics of PH patients and estimated burden of the disease in the population.

Methods
Prospective data collection took place in 5 University Hospitals (Basel, Bern, Geneva, Lausanne, Zurich) and in 4 associated hospitals (Aarau, Barmelweid, Locarno, St Gallen) in Switzerland. Patients with pulmonary arterial hypertension (idiopathic, associated with collagen vascular disease, portal hypertension, HIV infection, drugs or toxins), PH associated with lung diseases and/or hypoxaemia, PH due to chronic thrombotic and/or embolic disease, or PH due to miscellaneous disorders (sarcoidosis, histiocytosis X, lymphangiomyomatosis, compression of pulmonary vessels) were included in the registry. Patients with PH due to left heart disease were not included as well as three patients with pulmonary angiosarcoma. Due to a change in nomenclature in 2004, the term "idiopathic pulmonary arterial hypertension" (IPAH) includes both sporadic and familial cases of pulmonary arterial hypertension (PAH).
Pulmonary hypertension was defined as a mean pulmonary artery pressure of >25 mm Hg at rest or >30 mm Hg on exercise.
The following data were collected: age and gender of the patient, diagnosis according to the Evian clinical classification [6], date of diagnosis, height, weight, vital signs (blood pressure, heart rate and oxygen saturation at rest), NYHA class, six minute walking distance (6MWD), haemodynamics assessed by right heart catheterisation (systolic pulmonary artery pressure, mean pulmonary artery pressure (PAPmean), right atrial pressure, cardiac output, mixed venous oxygen saturation, pulmonary vascular resistance, response to vasoreactivity testing, tricuspid pressure gradient measured with transthoracic Doppler echocardiography, liver enzymes, creatinine, coagulation studies (INR) and current medication (diuretics, anticoagulation, calcium antagonists, iloprost, bosentan, sildenafil, oxygen). A positive response to vasoreactivity testing was defined as a decrease of PAPmean of more than 10 mm Hg, to a value lower than 40 mm Hg, with a normal or high cardiac output [7]. The tricuspid pressure gradient was calculated from the maximal transtricuspid jet in Doppler echocardiography using the modified Bernoulli equation.
Whether a patient was included in the registry or not was at discretion of the treating PH specialist. When a patient was included in the registry, data collection started at the first visit of the patient in the PH centre or associated hospital, and continued every 3 to 6 months until occurrence of one of the clinical endpoints (death, lung transplantation, pulmonary thrombendarterectomy). The clinical follow-up of the patients was performed as usual irrespective of whether a patient was included in the registry or not.
In this study, we analyse data from adult patients (>18 years of age) that were included in the registry between January 1999 and December 2004 (table 1).
Statistics: in descriptive analyses we report median [interquartile range] for continuous parameters and for categorical parameters frequency distribution.
For estimation of prevalence and incidence of PH in Switzerland we assumed an adult population of 6 million inhabitants [8]. The prevalence of PH for the year 2004 was calculated by dividing the number of patients diagnosed with PH and included in the registry before 2004, and known to be alive in 2004, by the approximate number of adult Swiss inhabitants. Incidence estimates were calculated for each year by dividing the number of patients diagnosed with PH and included in the registry each year by the approximate number of adult Swiss inhabitants (6 million).
For survival analyses, follow-up time was calculated from date of registration in the registry until last clinical visit or the date of death. Differences between the two outcome groups (survivors and non-survivors) were described by presenting Kaplan-Meier survival curves for selected variables. Analysis was performed with STATA 9.1 software (Stata Corp LP, Texas, USA).
Ethics: written informed consent was obtained from each patient before inclusion in the registry.  Table 1 Results A total of 252 adult patients were included in the registry between January 1999 and December 2004. Two patients with PH due to left heart disease were mistakenly included in the registry, and were excluded from further analysis. Median follow-up of the remaining 250 patients was 18.8    figure 1.
Seventy-one of the 250 patients had no rightheart catheter data in the registry (28%). Of the remaining 179 patients, 119 (66%) had vasoreactivity testing. A first analysis gave 38% of responders, which appeared much higher than reported in the literature. So right heart catheter data were verified in the source document in 118 of 250 patients. Of these 118 patients, 31 patients did not have a right-heart catheter, and 16 had a catheter but no vasoreactivity testing. In the remaining 71 patients, vasoreactivity was verified according to the criteria described above. New analysis gave an overall responder rate of 20% (table 2). Patients with IPAH were significantly more likely to re-spond to vasoreactivity testing than were patients with other aetiologies for PH (p = 0.03).
In 2004 we estimated the prevalence of PH of any aetiology in Switzerland at about 25 patients per one million adult inhabitants. For PAH prevalence was 15.5 patients, and for IPAH 8.6 patients per one million adult inhabitants respectively. The incidence of PAH was estimated to have increased from 1.2 patients per one million adult inhabitants in 1999 to 3.5 patients per one million adult inhabitants in 2004. Respective data for IPAH were 0.8 patients per one million adult inhabitants in 1999 and 1.2 patients per one million adult inhabitants in 2004.

Figure 2
Survival as a function of prognostically important baseline parameters.

Discussion
Analysis of the Swiss Registry for Pulmonary Hypertension yields an estimation of prevalence and incidence of this disease in Switzerland and gives some insights in the clinical characteristics of patients with PH. However, the registry has some limitations. Although comprehensive information has been provided to potential referring physicians all over Switzerland we cannot assume that every patient diagnosed with PH in Switzerland is included in the registry. Further, it is likely that patients with functional class I and II are under-represented probably in part because they are under-diagnosed. Moreover, the assessment of patients with PH in the different participating centres is not identical, and the completeness of data entered into the register varies.
The conservative estimates of the prevalence of PAH and IPAH in Switzerland in 2004 based on our registry are within the range of the data from the French Network on Pulmonary Arterial Hypertension with an estimated prevalence of 15 and 5.9 cases respectively per one million adult inhabitants [11]. Both the French and Swiss registries show a lower prevalence of this disease than the recently published data from the Scottish population [26]. Peacock and co-workers found a prevalence of PAH of 26 and of IPAH of 9 per million populations, based on data from the Scottish Pulmonary Vascular Unit. Their data from the national hospitalisation registry showed even higher values (52 and 25 per million, respectively). The conservatively estimated incidence of PAH in Switzerland in 2004 is in the range of the estimation based on the French registry (2.4 cases per one million adult inhabitants) [11], but again lower than the estimated incidence of 7.1 per million in the Scottish population [26]. Further studies from other countries have yet to show which estimates of prevalence and incidence are closest to reality, and whether regional differences play a role. Compared to retrospective data from 1991 to 1999 on PH in Switzerland, the proportion of patients >50 years of age (68%) is now much higher (figure 1) [5]. The reason for this might be the availability of several effective treatments for this disease [1][2][3] and the increasing clinical experience with those treatments. Physicians therefore consider this disease more often as a potential explanation for exercise intolerance also in elderly patients. The age shift is also due to the higher percentage of patients in the registry with CTEPH (29%) or hypoxaemia-related PH (7%) than documented in the past (15% and 0% respectively) [5]. The increase in the number of patients diagnosed with CTEPH may have been due to a major awareness following the publication of a study showing a more frequent than expected development of PH after acute pulmonary embolism [14].
Our patients are still detected at an advanced stage of the disease, with a mean pulmonary artery pressure of 50 mm Hg and 87% of patients having NYHA class III or IV dyspnoea at diagnosis. Reasons for this late detection might still be an insufficient awareness of the disease and the under-use of echocardiography and diffusion capacity measurements in the workup of patients with unexplained dyspnoea. It is also the hallmark of the disease itself, which progresses insidiously without specific clinical signs or symptoms until late, when right heart failure and severe disturbance of gas exchange is present.
Overall mortality over 5 years' follow-up of patients with PH of any aetiology included into the Swiss registry between 1999 and 2004 was 26%. There is no gender difference in mortality. In this study 6MWD, NYHA class, right atrial pressure and mixed venous saturation seem to be important prognostic parameters. There are other studies, including the retrospective data from 1991-1999 from the Swiss Registry for Pulmonary Hypertension [5] confirming the importance of central venous saturation [10], NYHA functional class [9,13], right atrial pressure [9,15] and 6-minute walking distance [12] as prognostic parameters. Pulmonary vascular resistance is neither a predictor of mortality in this study nor in others [5,12]. In contrast to other publications, parameters like mean pulmonary artery pressure, cardiac index, or response to vasoreactivity testing (15)(16)(17)(18) could not be identified as significant outcome parameters. Other parameters shown to have prognostic importance such as the presence and size of pericardial effusion on echocardiography [19], blood tests such as plasma brain natriuretic peptide (20), serum cardiac troponin T [21] or hyperuricaemia [22] and quality of life assessments [23] were not routinely collected in this registry.
The lack of prognostic significance of vasoreactivity testing might be due to the small number of patients who had vasoreactivity testing with verified response according to the Venice criteria (n = 71), with only 20 patients having idiopathic pulmonary arterial hypertension. The overall vasoreactivity rate of 20% is in accordance with other data in the literature [24]. However, right heart catheterisation and vasoreactivity testing remain a weak point of the routine assessment of patients with PH in Switzerland. Twenty-eight percent of patients did not undergo right heart catheterisation, and only 71% of patients who had right heart catheterisation for diagnosis had vasoreactivity testing. Whether this is a reporting bias or reflects reality is not known. It could be argued that in patients with CTEPH or PH due to hypoxaemia vasoreactivity, testing is not necessary, as the prognostic value of this manoeuvre has been validated only in patients with pulmonary arterial hypertension. However, vasoreactivity testing was performed in only 57% of patients with idiopathic pulmonary arterial hypertension, but in 44% of patients with CTEPH and 50% of patients with PH due to hypoxaemia or chronic lung disease. The significance of positive vasoreactivity test in the latter patients is not yet clear, but the registry might help to assess this point in the future -provided that right heart catheterisation and vasoreacitivity testing are consequently performed in every patient. As age and co-morbidities of patients with PH increase, the assessment of PH by right heart catheterisation is strongly recommended. It is the only method that reliably excludes left heart insufficiency as a reason for the elevated pulmonary artery pressure; moreover it may correct falsely elevated pressure gradients measured by echocardiography in patients with COPD [25].
In conclusion, the Swiss Registry for Pulmonary Hypertension is a useful and informative tool for clinicians involved in the management of patients with this disease and provides data on epidemiology of PH patients that compares well with data from other registries. A high reporting rate is crucial to have reliable information from the registry and to improve our epidemiological understanding of this disease. It may soon be important to identify patients at an earlier stage of disease to intervene earlier with new therapies. A sustained collaboration between general practitioners and specialised PH centres is necessary in this regard. The management of these patients in specialised centres guarantees a high quality of care, and the regular analysis of the registry data could be an instrument for quality control and might help identify weak points in assessment and treatment of patients affected with a disease leading to marked disabilities and still high mortality.
We thank Ulla Treder (University Hospital of Zurich) and Monika Sorge Maître (University Hospital of Geneva) for their help with the registry, and Markus Schwerzmann (Cardiology, University Hospital of Berne) for statistical support.