Szeto 2003.

Methods	Country: Hong Kong Setting/Design: University, randomised trial Time frame: 12 months Randomisation method: not stated Blinding ‐ Participants: Yes ‐ Investigators: Yes ‐ Outcome assessors: Yes ‐ Data analysis: Yes Intention‐to‐treat: Yes Loss to follow‐up: 4/30 and 3/30 in intervention and control respectively
Participants	INCLUSION CRITERIA Plasma bicarbonate level measured twice in the 3 months prior to enrolment. Total weekly Kt/V < 2.1 Venous bicarbonate ≤ 25 mmol/L on 2 consecutive measurements Stable clinical condition and CAPD regimen for at least 12 months TREATMENT GROUP Number: 30 Age: 54.3 ± 12.4 years Sex (M/F): 16/14 Duration of dialysis: 39.9 ± 20.8 months Diabetes: 12 CONTROL GROUP Number: 30 Age: 56.6 ± 13.2 years Sex (M/F): 19/11 Duration of dialysis: 39.4 ± 26.0 months Diabetes: 8 EXCLUSIONS Patients unlikely to survive, planned living‐related kidney transplant, transfer to other renal centre within 6 months
Interventions	TREATMENT GROUP Oral sodium bicarbonate 0.9 g thrice daily CONTROL GROUP Placebo (pure starch tablet) CO‐INTERVENTIONS None stated
Outcomes	Total Kt/V Residual GFR Subjective global assessment of nutrition (SGA) Normalised protein nitrogen appearance (NPNA) Anthropometric lean body mass (LBM) Anthropmetric measures (e.g. biceps skinfold thickness) Oedema, weight, hospitalisation, technique failure, mortality
Notes	Appearance, packaging and labeling of tablets were identical
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

iPTH ‐ immunoactive parathormone