Distinct phenotypes of hospitalized patients with hyperkalemia by machine learning consensus clustering and associated mortality risks.

BACKGROUND: Hospitalized patients with hyperkalemia are heterogeneous, and cluster approaches may identify specific homogenous groups. This study aimed to cluster patients with hyperkalemia on admission using unsupervised machine learning (ML) consensus clustering approach, and to compare characteristics and outcomes among these distinct clusters. METHODS: Consensus cluster analysis was performed in 5133 hospitalized adult patients with admission hyperkalemia, based on available clinical and laboratory data. The standardized mean difference was used to identify each cluster's key clinical features. The association of hyperkalemia clusters with hospital and 1-year mortality was assessed using logistic and Cox proportional hazard regression. RESULTS: Three distinct clusters of hyperkalemia patients were identified using consensus cluster analysis: 1661 (32%) in cluster 1, 2455 (48%) in cluster 2 and 1017 (20%) in cluster 3. Cluster 1 was mainly characterized by older age, higher serum chloride and acute kidney injury (AKI), but lower estimated glomerular filtration rate (eGFR), serum bicarbonate and hemoglobin. Cluster 2 was mainly characterized by higher eGFR, serum bicarbonate and hemoglobin, but lower comorbidity burden, serum potassium and AKI. Cluster 3 was mainly characterized by higher comorbidity burden, particularly diabetes and end-stage kidney disease, AKI, serum potassium, anion gap, but lower eGFR, serum sodium, chloride and bicarbonate. Hospital and 1-year mortality risk was significantly different among the three identified clusters, with highest mortality in cluster 3, followed by cluster 1 and then cluster 2. CONCLUSION: In a heterogeneous cohort of hyperkalemia patients, three distinct clusters were identified using unsupervised ML. These three clusters had different clinical characteristics and outcomes.

Gastrointestinal bleeding among hospitalizations for salicylate poisoning in the United States.

BACKGROUND: This study aimed to determine the incidence, as well as evaluate risk factors, and impact of gastrointestinal bleeding on outcomes and resource use in patients admitted for salicylate poisoning. METHODS: We used the National Inpatient Sample to construct a cohort of patients hospitalized primarily for salicylate poisoning from 2003 to 2014. We compared clinical characteristics, in-hospital treatments, outcomes and resource use between salicylate poisoning patients with and without gastrointestinal bleeding. RESULTS: Of 13 805 hospital admissions for salicylate poisoning, gastrointestinal bleeding occurred in 482 (3.5%) admissions. The risk factors for gastrointestinal bleeding included older age, history of atrial fibrillation and cirrhosis. After adjusting for difference in baseline characteristics, patients with gastrointestinal bleeding required more gastric lavage, gastrointestinal endoscopy, invasive mechanical ventilation and red blood cell transfusion. Gastrointestinal bleeding was significantly associated with increased risk of anemia, circulatory, liver and hematological failure but was not significantly associated with increased in-hospital mortality. The length of hospital stay and hospitalization cost was significantly higher in patients with gastrointestinal bleeding. CONCLUSION: Gastrointestinal bleeding occurred in about 4% of patients admitted for salicylate poisoning. Gastrointestinal bleeding was associated with higher morbidity and resource use but not mortality.

Nonsteroidal anti-inflammatory drugs and incidence of atrial fibrillation: a meta-analysis.

BACKGROUND: Nonsteroidal anti-inflammatory drugs (NSAIDs) are widely used for many inflammatory disorders and pain-related illnesses. Despite their widespread use, the association between NSAIDs and the incidence of atrial fibrillation (AF) remains unclear. The aim of this systematic review and meta-analysis is to investigate this association. METHODS: A systematic review was conducted in MEDLINE, EMBASE and Cochrane databases from inception through August 2019 to identify studies that evaluated the risk of AF among patients using NSAIDs. Pooled risk ratios (RRs) and 95% CI were calculated using a random-effect, generic inverse variance method. The protocol for this meta-analysis is registered with PROSPERO (International Prospective Register of Systematic Reviews; no. CRD42019141609). RESULTS: Eight observational studies (four case-control studies and four cohort studies) with a total of 14 806 420 patients were enrolled. When compared with nonNSAIDs users, the pooled RR of AF in patients with NSAIDs use was 1.29 (95% CI 1.19-1.39). Meta-analyses based on the type of study were additionally performed. Subgroup analysis by study design revealed a significant association between the use of NSAIDs and AF for both case-control studies (pooled RR 1.37; 95% CI, 1.15-1.63) and cohort studies (pooled RR 1.22; 95% CI, 1.14-1.31). Sub-analyses based on specific NSAIDs showed pooled RRs of AF in patients using ibuprofen of 1.30 (95% CI 1.22-1.39), naproxen of 1.44 (95% CI 1.18-1.76) and diclofenac of 1.37 (95% CI 1.10-1.71), respectively. Funnel plot and Egger's regression asymmetry tests were performed and showed no publication bias. CONCLUSION: NSAID use is associated with incident AF. Our study also demonstrated a consistent result among different NSAIDs.

Impacts of admission serum albumin levels on short-term and long-term mortality in hospitalized patients.

AIM: The aim of this study is to assess the association between admission serum albumin and short- and long-term mortality in all hospitalized patients. DESIGN: A single-center cohort study. METHODS: A retrospective cohort of all adult hospitalized patients at a tertiary referral hospital between January 2009 and December 2013 were analysed. Admission serum albumin was stratified into six groups: ≤2.4, 2.5-2.9, 3.0-3.4, 3.5-3.9, 4.0-4.4 and ≥4.5 g/dl. The outcomes of interest were in-hospital mortality, length of hospital stay and 1-year mortality. Serum albumin of 4-4.4 g/dl was selected as a reference group for outcome comparison. RESULTS: A total of 14 075 patients were studied. Admission serum albumin of ≥4.5 g/dl had the lowest in-hospital and 1-year mortality with progressively increased in-hospital mortality observed with decreased admission serum albumin. In adjusted analysis, compared with serum albumin of 4.0-4.4 g/dl, serum albumin of ≤2.4, 2.5-2.9, 3.0-3.4 and 3.5-3.9 were significantly associated with increased in-hospital and 1-year mortality. In contrast, serum albumin of ≥4.5 g/dl was significantly associated with lower 1-year mortality but not in-hospital mortality. Admission serum albumin <4.0 g/dl was significantly associated with a prolonged hospital stay, while admission serum albumin of ≥4.5 g/dl was significantly associated with shorter hospital stay, compared with serum albumin of 4.0-4.4 g/dl. CONCLUSION: Low albumin level at admission was progressively associated with increased short- and long-term mortality in all hospitalized patients even when albumin level was considered in normal range.

Effects of aspirin and non-steroidal anti-inflammatory drugs on the risk of cholangiocarcinoma: a meta-analysis.

BACKGROUND: Non-steroidal anti-inflammatory drugs (NSAIDs) can suppress the proliferation of cholangiocarcinoma (CCA) cells in vitro through inhibition of cyclooxygenase-2. However, the effects of aspirin and NSAIDs on the risk of CCA remain unclear. We performed this meta-analysis to assess the risk of biliary tract cancers in patients who take aspirin and/or NSAIDs. METHODS: A systematic review was conducted utilizing MEDLINE, EMBASE, Cochrane databases from inception through October 2017 to identify studies that assessed the association of aspirin and/or NSAIDs use with risk of biliary tract cancers including CCA, gallbladder cancer and ampulla of Vater cancer. Effect estimates from the studies were extracted and combined using the random-effect, generic inverse variance method of DerSimonian and Laird. RESULTS: Five observational studies with a total of 9 200 653 patients were enrolled. The pooled OR of CCA in patients with aspirin use was 0.56 (95% CI, 0.32-0.96). Egger's regression asymmetry test was performed and showed no publication bias for the association between aspirin use and CCA with P = 0.42. There was no significant association between NSAIDs use and CCA, with a pooled OR of 0.79 (95% CI, 0.28-2.21). One study showed a significant association between aspirin use and reduced risk of gallbladder cancer with OR of 0.37 (0.17-0.80). However, there was no significant association between aspirin and ampulla of Vater cancer with OR of 0.22 (0.03-1.65). CONCLUSIONS: Our study demonstrates a significant association between aspirin use and a 0.56-fold decreased risk of CCA. However, there is no association between the use of NSAIDs and CCA.

Impact of admission serum potassium on mortality in patients with chronic kidney disease and cardiovascular disease.

BACKGROUND: Little is known about the effect of admission potassium (K) on risk of in-hospital mortality in chronic kidney disease (CKD) and cardiovascular disease (CVD) patients. AIM: The aim of this study was to assess the relationship between admission serum K and in-hospital mortality in all hospitalized patients stratified by CKD and/or CVD status. DESIGN AND METHODS: All adult hospitalized patients who had admission serum K between years 2011 and 2013 were enrolled. Admission serum K was categorized into seven groups (<3.0, 3.0-3.5, 3.5-4.0, 4.0-4.5, 4.5-5.0, 5.0-5.5 and ≥5.5 mEq/L). The odds ratio (OR) of in-hospital mortality by admission serum K, using K 4.0-4.5 mEq/L as the reference group, was obtained by logistic regression analysis. RESULTS: 73,983 patients were studied. The lowest incidence of in-hospital mortality was associated with serum K within 4.0-4.5 mEq/L. A U-shaped curve emerged demonstrating higher in-hospital mortality associated with both serum K < 4.0 and >4.5 mEq/L. After adjusting for potential confounders, both serum K < 4.0 mEq/L and >5.0 mEq/L were associated with increased in-hospital mortality with ORs of 3.26 (95% CI 2.03-4.98), 2.40 (95% CI 1.89-3.04), 1.38 (95%CI 1.15-1.66), 1.89 (95% CI 1.49-2.38) and 3.62 (95%CI 2.73-4.76) when serum K were within <3.0, 3.0-3.5, 3.5-4.0, 5.0-5.5, and ≥5.5 mEq/L, respectively. In CVD patients, the highest in-hospital mortality was associated with serum K < 3.0 mEq/L (OR 1.70, 95%CI 1.31-2.18). In CKD patients, the highest in-hospital mortality was associated with serum K ≥ 5.5 mEq/L (OR 3.26, 95%CI 2.14-4.90). CONCLUSION: Admission serum K < 4.0 mEq/L and >5.0 mEq/L were associated with increased in-hospital mortality. The mortality risk among patients with various admission potassium levels was affected by CKD and/or CVD status.

Hypomagnesaemia linked to depression: a systematic review and meta-analysis.

BACKGROUND: The reported risk of depression in patients with hypomagnesaemia is controversial. AIM: The objective of this meta-analysis was to assess the association between depression and hypomagnesaemia. METHODS: A literature search was performed using MEDLINE, EMBASE, Cochrane Database and clinicaltrials.gov from inception through October 2014. Studies that reported odds ratios, relative risks or hazard ratios comparing the risk of depression in patients with hypomagnesaemia were included. Pooled risk ratios (RR) and 95% confidence interval (CI) were calculated using a random-effect, generic inverse variance method. RESULTS: Six observational studies (three cohort studies, two cross-sectional studies and a case-control study) with a total of 19,137 patients were identified and included in the data analysis. The pooled RR of depression in patients with hypomagnesaemia was 1.34 (95% CI, 1.01-1.79, I(2) = 33%). The association between depression and hypomagnesaemia was marginally insignificant after the sensitivity analysis including only cohort and case-control studies, with a pooled RR of 1.38 (95% CI, 0.92-2.07, I(2) = 24%). CONCLUSION: Our study demonstrates a potential association between hypomagnesaemia and depression. Further studies assessing the benefits of treatment of hypomagnesaemia in patients with depression are needed.