Discussion
In the present study 123 patients with ESRD were followed up for one year to evaluate comorbidities that required hospitalization in association with epidemiological characteristics such as gender, age, diabetes, mode of dialysis and admissions in the hospital (a marker of overall morbidities or well being). The mortality in the same cohort was evaluated during a longer period of 33 months. The percent rates for the major causes of ESRD in our prevalent population in 2009 was very similar to the average rates reported for the whole of Europe by the ERA/EDTA registry 2009 annual report.17 In particular, the major cause of ESRD was diabetic nephropathy (19.5%), a percentage that is much lower than that reported (33%) for the United States.18 It is of interest that a much higher proportion of patients are receiving PD in our center (37.4%) as compared to the rest of Greece (8.6%) as well as several other European countries (4-14%). This finding may be related to the mountain terrain of our island that impedes easy approach to the hospital based hemodialysis units of the urban areas. In our prevalent HD patients, 75.3% had a functioning AVF, 15.7% a functioning AVG and only 9% carried a jugular catheter which is in accordance to the European standards.19 Despite the low percentage of central catheters in our population, catheter related infections comprised a major cause of hospital admissions. This finding indicates the need for further reduction in catheter usage with a careful surveillance and early repair of problematic fistulas. Another major cause of hospital admissions was related to cardiac morbidity, in particular pulmonary edema. Therefore a more careful assessment of patients’ dry weight could substantially reduce admission rates and morbidity. Admission rates and hospital stay length in our cohort appear improved than other centers, a finding that is difficult to interpret especially when considering our population’s relatively advanced age and long dialysis vintage.20,21 Interestingly, patients with longer dialysis vintage had fewer hospitalizations and shorter hospital stay compared to patients who are new on dialysis. This finding implies that patients may need some time for their medical condition to stabilize after starting dialysis as well as for several associated problems, such as anemia and hypertension, to subside. An alternative explanation could be that during the first years of dialysis the patients with severe co-morbidities (and highest hospitalization rates) die, while healthier patients with fewer hospitalizations survive longer.22 In support of this statement, the mortality curves shown in figure 3, for both HD and PD, were steeper during the first 50 months than afterwards. Therefore early years on dialysis are burdened with a greater morbidity and mortality risk. Another worth mentioning finding of this study concerns the mortality difference between HD and PD. It was apparent that after the first 2 years the two curves deviate in favor of HD. This finding should be seen with caution though, since some comorbidities were not taken into account in this study. For example in our center, patients with severe heart failure are usually treated with PD rather than HD, a practice that could negatively affect the survival rates in our PD group. Finally, the study population consisted of patients on PD or HD treated in one hospital, therefore the findings of the study should be viewed under this limitation.