Limits...
The association between the use of proton pump inhibitors and the risk of hypomagnesemia: a systematic review and meta-analysis.

Park CH, Kim EH, Roh YH, Kim HY, Lee SK - PLoS ONE (2014)

Bottom Line: Pooled odds ratios with 95% confidence intervals were calculated using the random effects model.On meta-analysis, pooled odds ratio for PPI use was found to be 1.775 (95% confidence interval 1.077-2.924).However, significant heterogeneity among the included studies prevented us from reaching a definitive conclusion.

View Article: PubMed Central - PubMed

Affiliation: Division of Gastroenterology, Department of Internal Medicine, Severance Hospital, Institute of Gastroenterology, Yonsei University College of Medicine, Seoul, Korea.

ABSTRACT

Background: Although many case reports have described patients with proton pump inhibitor (PPI)-induced hypomagnesemia, the impact of PPI use on hypomagnesemia has not been fully clarified through comparative studies. We aimed to evaluate the association between the use of PPI and the risk of developing hypomagnesemia by conducting a systematic review with meta-analysis.

Methods: We conducted a systematic search of MEDLINE, EMBASE, and the Cochrane Library using the primary keywords "proton pump," "dexlansoprazole," "esomeprazole," "ilaprazole," "lansoprazole," "omeprazole," "pantoprazole," "rabeprazole," "hypomagnesemia," "hypomagnesaemia," and "magnesium." Studies were included if they evaluated the association between PPI use and hypomagnesemia and reported relative risks or odds ratios or provided data for their estimation. Pooled odds ratios with 95% confidence intervals were calculated using the random effects model. Statistical heterogeneity was assessed with Cochran's Q test and I2 statistics.

Results: Nine studies including 115,455 patients were analyzed. The median Newcastle-Ottawa quality score for the included studies was seven (range, 6-9). Among patients taking PPIs, the median proportion of patients with hypomagnesemia was 27.1% (range, 11.3-55.2%) across all included studies. Among patients not taking PPIs, the median proportion of patients with hypomagnesemia was 18.4% (range, 4.3-52.7%). On meta-analysis, pooled odds ratio for PPI use was found to be 1.775 (95% confidence interval 1.077-2.924). Significant heterogeneity was identified using Cochran's Q test (df = 7, P<0.001, I2 = 98.0%).

Conclusions: PPI use may increase the risk of hypomagnesemia. However, significant heterogeneity among the included studies prevented us from reaching a definitive conclusion.

Show MeSH
Forest plots for risk of hypomagnesemia.PPI, proton pump inhibitor; CI, confidence interval.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4230950&req=5

pone-0112558-g002: Forest plots for risk of hypomagnesemia.PPI, proton pump inhibitor; CI, confidence interval.

Mentions: The cut-off value of serum magnesium level for defining hypomagnesemia was 1.6, 1.7, and 1.8 mg/dL in one [25], five [22], [24], [26], [28], [29], and three studies [23], [27], [30], respectively (Table 2). Among patients taking PPIs, the median proportion of patients with hypomagnesemia in all included studies was 27.1% (range, 11.3–55.2%). In addition, the median of the proportion of patients with hypomagnesemia in those not taking PPIs was 18.4% (range, 4.3–52.7%) across the studies. Of 9 included studies, six reported both unadjusted OR and adjusted OR. Other two studies showed unadjusted OR only. Remaining one study showed adjusted OR only. Most studies adjusted for the following confounders: use of diuretics (6/7), renal function (5/7), age (4/7), diabetes mellitus (4/7), and comorbidities (4/7). On meta-analysis, pooled unadjusted OR for PPI use was found to be 1.775 (95% CI = 1.077–2.924). Significant heterogeneity was identified (Cochran’s Q test, df = 7, P<0.001, I2 = 98.0%). This risk increase with PPI use persisted even after adjusting for potential confounders where reported in studies (pooled adjusted OR [95% CI] = 1.484 [1.103–1.997], Fig. 2), although the heterogeneity persisted (Cochran’s Q test, df = 6, P<0.001, I2 = 89.1%).


The association between the use of proton pump inhibitors and the risk of hypomagnesemia: a systematic review and meta-analysis.

Park CH, Kim EH, Roh YH, Kim HY, Lee SK - PLoS ONE (2014)

Forest plots for risk of hypomagnesemia.PPI, proton pump inhibitor; CI, confidence interval.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4230950&req=5

pone-0112558-g002: Forest plots for risk of hypomagnesemia.PPI, proton pump inhibitor; CI, confidence interval.
Mentions: The cut-off value of serum magnesium level for defining hypomagnesemia was 1.6, 1.7, and 1.8 mg/dL in one [25], five [22], [24], [26], [28], [29], and three studies [23], [27], [30], respectively (Table 2). Among patients taking PPIs, the median proportion of patients with hypomagnesemia in all included studies was 27.1% (range, 11.3–55.2%). In addition, the median of the proportion of patients with hypomagnesemia in those not taking PPIs was 18.4% (range, 4.3–52.7%) across the studies. Of 9 included studies, six reported both unadjusted OR and adjusted OR. Other two studies showed unadjusted OR only. Remaining one study showed adjusted OR only. Most studies adjusted for the following confounders: use of diuretics (6/7), renal function (5/7), age (4/7), diabetes mellitus (4/7), and comorbidities (4/7). On meta-analysis, pooled unadjusted OR for PPI use was found to be 1.775 (95% CI = 1.077–2.924). Significant heterogeneity was identified (Cochran’s Q test, df = 7, P<0.001, I2 = 98.0%). This risk increase with PPI use persisted even after adjusting for potential confounders where reported in studies (pooled adjusted OR [95% CI] = 1.484 [1.103–1.997], Fig. 2), although the heterogeneity persisted (Cochran’s Q test, df = 6, P<0.001, I2 = 89.1%).

Bottom Line: Pooled odds ratios with 95% confidence intervals were calculated using the random effects model.On meta-analysis, pooled odds ratio for PPI use was found to be 1.775 (95% confidence interval 1.077-2.924).However, significant heterogeneity among the included studies prevented us from reaching a definitive conclusion.

View Article: PubMed Central - PubMed

Affiliation: Division of Gastroenterology, Department of Internal Medicine, Severance Hospital, Institute of Gastroenterology, Yonsei University College of Medicine, Seoul, Korea.

ABSTRACT

Background: Although many case reports have described patients with proton pump inhibitor (PPI)-induced hypomagnesemia, the impact of PPI use on hypomagnesemia has not been fully clarified through comparative studies. We aimed to evaluate the association between the use of PPI and the risk of developing hypomagnesemia by conducting a systematic review with meta-analysis.

Methods: We conducted a systematic search of MEDLINE, EMBASE, and the Cochrane Library using the primary keywords "proton pump," "dexlansoprazole," "esomeprazole," "ilaprazole," "lansoprazole," "omeprazole," "pantoprazole," "rabeprazole," "hypomagnesemia," "hypomagnesaemia," and "magnesium." Studies were included if they evaluated the association between PPI use and hypomagnesemia and reported relative risks or odds ratios or provided data for their estimation. Pooled odds ratios with 95% confidence intervals were calculated using the random effects model. Statistical heterogeneity was assessed with Cochran's Q test and I2 statistics.

Results: Nine studies including 115,455 patients were analyzed. The median Newcastle-Ottawa quality score for the included studies was seven (range, 6-9). Among patients taking PPIs, the median proportion of patients with hypomagnesemia was 27.1% (range, 11.3-55.2%) across all included studies. Among patients not taking PPIs, the median proportion of patients with hypomagnesemia was 18.4% (range, 4.3-52.7%). On meta-analysis, pooled odds ratio for PPI use was found to be 1.775 (95% confidence interval 1.077-2.924). Significant heterogeneity was identified using Cochran's Q test (df = 7, P<0.001, I2 = 98.0%).

Conclusions: PPI use may increase the risk of hypomagnesemia. However, significant heterogeneity among the included studies prevented us from reaching a definitive conclusion.

Show MeSH