ASN's Mission

ASN leads the fight to prevent, treat, and cure kidney diseases throughout the world by educating health professionals and scientists, advancing research and innovation, communicating new knowledge, and advocating for the highest quality care for patients.

learn more

Contact ASN

1401 H St, NW, Ste 900, Washington, DC 20005


The Latest on Twitter

Kidney Week

Abstract: PO1176

Is an Increase in Anion Gap a Predictor of Hemodialysis Initiation in Patients with Advanced CKD?

Session Information

Category: Fluid, Electrolyte, and Acid-Base Disorders

  • 902 Fluid, Electrolyte, and Acid-Base Disorders: Clinical


  • Kukimoto, Hikaru, Kyorin Daigaku, Mitaka, Tokyo, Japan
  • Fukuoka, Kazuhito, Kyorin Daigaku, Mitaka, Tokyo, Japan
  • Kawashima, Soko, Kyorin Daigaku, Mitaka, Tokyo, Japan
  • Ikegaya, Noriko, Kyorin Daigaku, Mitaka, Tokyo, Japan
  • Kawakami, Takahisa, Kyorin Daigaku, Mitaka, Tokyo, Japan
  • Kishimoto, Mitsumasa, Kyorin Daigaku, Mitaka, Tokyo, Japan
  • Komagata, Yoshinori, Kyorin Daigaku, Mitaka, Tokyo, Japan
  • Kaname, Shinya, Kyorin Daigaku, Mitaka, Tokyo, Japan

Because uremic symptoms and manifestations vary among patients with advanced chronic kidney disease, it is sometimes difficult to decide on the timing of dialysis initiation only from them. Thus, we attempted to investigate whether anion gap (AG) that may reflect the accumulation of total organic acids in uremia can be a marker of uremia and may predict the timing of dialysis initiation.


This study included pre-dialysis patients who attended to our hospital for more than six months prior to the beginning of hemodialysis (HD), and retrospectively analyzed the relationship between their serological data, AG, and various uremic symptoms. The AG was calculated as the corrected AG (cAG) = Na-Cl-HCO3 [nmol/L]+2.5 x (4-serum albumin concentration [g/L]). The statistical analysis was performed by logistic regression analysis, correlation analysis, and factor analysis using SPSS®.


A total of 283 patients [diabetes mellitus: 136 (48.1%), nephrosclerosis: 66 (23.3%), glomerulonephritis: 36 (12.7%)] were included in this study. The most common clinical symptom before dialysis initiation was fluid overload, which was seen in 134 patients (47.3%), followed by anorexia 104 patients (36.7%) and general malaise 96 patients (33.9%). The cAG began to increase 3 months before the initiation of HD (14.2 mmo/L), which showed a rapid increase just before the initiation, and was correlated with anorexia and fatigue, better than fluid retention. Of note is that cAG was most significantly associated with dialysis initiation among various factors. The ROC of cAG for dialysis initiation showed the highest value of AUC 0.797 (95% CI=0.72 to 0.85, p=0.05), with a cutoff value of adjusted cAG 15.975 (sensitivity 0.689, specificity 0.786).


Uremic symptoms and some serological markers including azotemia, metabolic acidosis, and hyperphosphatemia have been usually used to predict the magnitude of uremia and the timing of dialysis initiation. In our study, it is suggested that a rapid increase in cAG over 16 mEq/L may also be a good predictor of dialysis initiation within the subsequent 3 months.