Age Differences in Associations of Serum Alkaline Phosphatase and Mortality Among Peritoneal Dialysis Patients

Introduction
Serum alkaline phosphatase (ALP) serves as a critical bone turnover marker in chronic kidney disease (CKD) patients. Prior studies have established a link between elevated ALP levels and increased mortality in hemodialysis (HD) patients. In peritoneal dialysis (PD) populations, however, the predictive value of ALP for mortality remains inconsistent. While some studies, including earlier work by this research group, identified ALP as an independent predictor of all-cause and cardiovascular mortality in PD patients, a conflicting report from Taiwan found no significant association over a 5-year follow-up period. This discrepancy highlights the need to clarify the role of ALP in PD outcomes and investigate potential modifying factors such as age.

This retrospective cohort study examined the relationship between serum ALP levels and all-cause mortality in younger (<65 years) and elderly (≥65 years) PD patients, hypothesizing that age modifies the prognostic significance of ALP. By analyzing longitudinal data from a large single-center PD population, the study aimed to resolve existing uncertainties and provide age-specific insights into ALP-associated mortality risks.

Methods
The study included 1,273 incident PD patients from The First Affiliated Hospital of Sun Yat-sen University, China, who initiated PD between January 2006 and December 2011. Key inclusion criteria were age ≥18 years and PD treatment duration >90 days. Patients with malignancies, failed renal transplants, or prior permanent HD were excluded. Baseline demographic, clinical, and laboratory data were collected during the first 3 months of PD initiation.

Serum ALP levels were measured using standardized laboratory methods. Patients were stratified into younger (n=1,068) and elderly (n=205) cohorts using a 65-year age threshold. Each age group was further divided into ALP quartiles:

• Younger patients: Q1 (≤56 U/L), Q2 (56–69 U/L), Q3 (70–88 U/L), Q4 (≥88 U/L)
• Elderly patients: Q1 (≤61 U/L), Q2 (61–76 U/L), Q3 (77–97 U/L), Q4 (≥97 U/L)

Outcomes were tracked until December 2013, with mortality as the primary endpoint. Survival analysis used Kaplan-Meier curves and log-rank tests. Cox proportional hazards models adjusted for covariates including age, diabetes, cardiovascular disease (CVD), residual urine output, serum albumin, hemoglobin, inflammatory markers (neutrophil-to-lymphocyte ratio), bone metabolism parameters (calcium, phosphorus, parathyroid hormone), and medication use (vitamin D analogs, phosphate binders).

Results
Baseline Characteristics
Younger patients (mean age 43.4±12.0 years) demonstrated lower ALP levels (median 70 U/L vs. 77 U/L in elderly), higher residual kidney function (24-hour urine output 1,000 mL vs. 800 mL), and better nutritional status (serum albumin 35.6 vs. 33.2 g/L) compared to the elderly cohort (mean age 71.8±4.6 years). The elderly group had higher comorbidity burdens, with significantly greater prevalence of diabetes (43.9% vs. 22.7%) and CVD (59.5% vs. 19.3%).

ALP Distribution
In younger patients, baseline ALP ranged from 10–933 U/L (11.9% exceeding the normal upper limit of 110 U/L). Higher ALP quartiles correlated with older age, diabetes, elevated liver enzymes (ALT, AST), and higher parathyroid hormone levels. Among elderly patients, ALP levels (15–389 U/L, 17.6% >110 U/L) showed no consistent patterns with other clinical parameters except lower systolic blood pressure in the highest quartile.

Mortality Outcomes
Over a median follow-up of 34.2 months (IQR 20.8–48.5):

• Younger cohort: 144 deaths (13.5%), 53.5% cardiovascular
• Elderly cohort: 104 deaths (50.7%), 55.8% cardiovascular

Kaplan-Meier Survival Analysis
Younger patients in the highest ALP quartile (Q4) showed significantly reduced survival versus lower quartiles (p=0.014):

• Q1: 1-/3-/5-year survival 97.0%, 92.6%, 90.7%
• Q4: 1-/3-/5-year survival 96.9%, 90.0%, 81.1%

In contrast, elderly patients exhibited no survival differences across ALP quartiles (p=0.683), with 5-year mortality rates ranging from 51.9%–54.9%.

Cox Regression Models
After full adjustment for confounders:

• Younger patients: Each 10 U/L ALP increase conferred 6.4% higher mortality risk (HR 1.064, 95% CI 1.03–1.10; p<0.001). Q4 patients had double the mortality hazard of Q1 (HR 2.01, 95% CI 1.07–3.75).
• Elderly patients: No significant ALP-mortality association (HR 1.04 per 10 U/L, 95% CI 0.99–1.10; p=0.101).

Discussion
Mechanistic Insights
The age-dependent ALP-mortality relationship may stem from differences in disease pathophysiology:

1. Vascular Calcification: ALP hydrolyzes pyrophosphate, a potent inhibitor of vascular calcification. Younger patients with preserved renal function may experience accelerated calcification at higher ALP levels, while elderly patients already exhibit advanced calcification from aging and comorbidities.
2. Inflammation and Nutrition: Elderly patients demonstrated lower albumin, higher inflammation (elevated neutrophil-to-lymphocyte ratio), and greater diabetes prevalence—factors that dominate mortality risk and overshadow ALP’s contribution.
3. Bone Metabolism: Stronger ALP-bone turnover coupling in younger patients links elevated ALP to mineral disorders, whereas elderly patients may have ALP elevations driven by non-skeletal sources like liver dysfunction.

Clinical Implications
These findings suggest ALP holds greater prognostic value in younger PD populations. Regular ALP monitoring and interventions to lower levels (e.g., optimizing mineral metabolism, reducing inflammation) could improve outcomes in this subgroup. For elderly patients, comprehensive management of comorbidities and nutritional status may outweigh ALP-focused strategies.

Limitations

1. Retrospective Design: Potential unmeasured confounders like vascular calcification scores or detailed medication adherence.
2. Single-Center Data: Limits generalizability despite the large sample size.
3. ALP Isoforms: Total ALP measurement without differentiation between bone, liver, or vascular isoforms.

Future Directions
Prospective studies should:

• Measure ALP isoforms and direct vascular calcification markers
• Explore age-specific ALP thresholds for clinical decision-making
• Investigate whether ALP-lowering therapies improve survival in younger PD patients

Conclusion
This study resolves prior inconsistencies by demonstrating that serum ALP predicts all-cause mortality exclusively in younger PD patients. The absence of association in elderly individuals underscores the need for age-specific risk stratification in PD populations. Clinicians should consider ALP as a modifiable risk factor for younger patients while prioritizing comorbidity management in older adults.

doi.org/10.1097/CM9.0000000000000019

Was this helpful?

YesNo

0 / 0