The survival of patients undergoing peritoneal dialysis (PD) has improved over the past decade, but their mortality rate remains high. The aim of the current study was to identify correctable and uncorrectable factors influencing survival according to the elapsed time in patients undergoing PD. We retrospectively analyzed data from medical records of 118 patients who had undergone PD for >6 months. We analyzed laboratory findings at three time points (point of PD initiation, 6-month point of PD and 3-month point prior to death or last follow-up) during PD treatment and prescribed medications taken for >50% of the follow-up period. Three-year survival group was younger, had lower prevalence rates of ischemic heart disease (p = 0.024) and heart failure (HF) (8.5% vs. 34.6%, p = 0.003), higher serum albumin levels (albumin 2) at the 6-month point of PD, and higher serum albumin (albumin3) and creatinine (creatinine3) levels at the 3-month point prior to death or last follow-up than nonsurvival group. Patients without underlying HF, patients treated with angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers during the last 6 months (p = 0.0042), and those having creatinine 3 >9.5 mg/dL (p = 0.0029), and albumin 2 (p = 0.0209) and albumin 3 >3.5 g/dL (p = 0.0003) showed significantly higher survival curves. HF and albumin 3 were found to be independent factors for 3-year survival and long-term survival, respectively, by the multivariate Cox regression model. In conclusion, HF was useful information for predicting 3-year survival, and low serum albumin levels according to elapsed time should be corrected for survival of PD patients.