Study population
We included 22 consecutive patients (Table ). One patient had to be excluded because she could not perform the examinations after week 3 and week 15, due to upper airway infection. Thus, the final study group consisted of 21 patients: 9 patients (43%) with SSc-APAH, 7 (33%) with SLE-APAH, 2 (10%) with MCTD and 3 (14%) with other CTDs such as Jo-1 syndrome and Sjöegren's syndrome. Demographic data, diagnosis, functional class, hemodynamic values, lung function and medical therapy in the full study population are summarized in Table . At baseline 9 patients (43%) were in WHO-FC II, 7 (33%) were classified as WHO-FC III and 5 (24%) as WHO-FC IV. Combination therapies, including two to three PAH-targeted agents, were used in 62% of patients (Table ).
Assessment of training effects
ET significantly improved the 6MWD from 386 ± 121 (mean ± standard deviation (SD)) meters by 64 ± 47 meters after 3 weeks (P < 0.001) and by 71 ± 35 meters after 15 weeks (P < 0.003) (Figure ). All patients except one improved their 6MWD (Figure ). In the patient with no increase in 6MWD the test was limited by her hip joint osteoarthritis. However, in this patient other parameters of physical exercise capacity improved during the ET.
Nine patients (43%) referred from PH-centers other than Heidelberg did not attend the visit after 15 weeks, mainly due to the long traveling distance. Results remained significant after multiple imputation of missing values for the 6MWD at 15 weeks using the explanatory variables as described in Methods. The improvement in 6MWD ranged from 71 to 79 meters in the imputation replications. Patients who did not attend the assessment after 15 weeks had a significantly higher peak oxygen consumption at the AT (716 ± 139 vs. 543 ± 161 ml/min; P = 0.038) and a higher improvement in 6MWD (79 ± 62 vs. 50 ± 24 meters; P = 0.156) after 3 weeks. They did not differ significantly in other parameters such as demographics or parameters of right heart catheterization. After Bonferroni adjustment, improvement in 6MWD remained statistically significant.
ET also significantly improved QoL parameters indicated by the SF-36 subscale scores for physical functioning (P = 0.025), general health perception (P = 0.049), social functioning (P = 0.008), mental health (P = 0.033) and vitality (P = 0.021) (Figure , Table ).
Mean peak oxygen consumption, peak oxygen consumption/kg body weight, oxygen consumption at the anaerobic threshold, and oxygen saturation at maximal workload during cardiopulmonary exercise testing increased significantly from baseline to 3 weeks and to 15 weeks (Table ). Heart rate at rest significantly decreased after 3 and 15 weeks. After 15 weeks patients achieved significantly higher workloads with an increase of maximal heart rate during cardiopulmonary exercise testing and maximal systolic blood pressure. The Borg scale remained unchanged although significantly higher workloads and higher heart rates during exercise were attained (Table Figure ). After 3 weeks of ET, mean diastolic blood pressure and sPAP at rest were significantly reduced (Table ). C-reactive protein reduced in trend, but not significantly, after 3 and 15 weeks, whereas N-terminal prohormone of brain natriuretic peptide (NT-proBNP) plasma-levels remained unchanged. Seven patients improved their WHO-FC after 15 weeks; five of them were reclassified as WHO-FC IV. Nevertheless, the change in WHO-FC after 3 weeks and after 15 weeks compared to baseline was not significant (P = 0.058 and P = 0.096, respectively), possibly due to the small sample size.
Adverse events
During the 3-week in-hospital training, three patients had adverse events as follows: gastrointestinal infection with diarrhea (n = 1) and respiratory infections (n = 2). In all three patients the infection was treated successfully so that they were able to continue the training program after one or two weeks. All other patients (81%) tolerated the ET well. There were no signs of clinical worsening or right heart failure during the in-hospital program.
Follow-up and survival
Follow-up data were obtained after 2.9 ± 1.9 years (Figure ). One patient had been listed for lung transplantation before starting the rehabilitation program. She improved her walking distance by about 88 meters after 3 weeks, but did not perform the examination after 15 weeks due to the long traveling distance. However, she required lung transplantation 6 months later. Between 2.8 and 5.0 years after in-hospital rehabilitation, two patients died due to PAH and right heart failure, and one patient died due to cancer of unknown primary origin. The Kaplan-Meier PAH-survival rate was 100% after 1 and 2 years, and 80% after 3 years (Figure ). The overall survival rate was 100% after 1 and 2 years, and 73% after 3 years. The transplantation-free survival rate was 95% after 1 and 2 years (Figure ).
ET was continued over 3 years by 11 patients (61%). They reported a mean ET-duration of 29 ± 18 minutes at 4 ± 2 days/week. Among the 11 patients, 44% continued bicycle ergometer training, 17% dumbbell-training, 33% walking and 22% alternative training, such as gymnastics. Four patients (22%) combined two training items and three patients (17%) combined three items. Six patients (33%) stopped exercise training for the following reasons: pain, presyncopes, no permanent instruction and confidence with general state of health.