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.

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.

Admission serum magnesium levels and the risk of acute respiratory failure

BACKGROUND

The association between admission serum magnesium (Mg) levels and risk of acute respiratory failure (ARF) in hospitalised patients is limited. The aim of this study was to assess the risk of developing ARF in all hospitalised patients with various admission Mg levels.

METHODS

This is a single-center retrospective study conducted at a tertiary referral hospital. All hospitalised adult patients who had admission Mg available from January to December 2013 were analysed in this study. Admission Mg was categorised based on its distribution into six groups (less than 1.5, 1.5-1.7, 1.7-1.9, 1.9-2.1, 2.1-2.3 and greater than 2.3 mg/dl). The primary outcome was in-hospital ARF occurring after hospital admission. Logistic regression analysis was performed to obtain the odds ratio of ARF of various admission Mg levels using Mg of 1.7-1.9 mg/dl as the reference group.

RESULTS

Of 9780 patients enrolled, ARF occurred in 619 patients (6.3%). The lowest incidence of ARF was when serum Mg within 1.7-1.9 mg/dl. A U-shaped curve emerged demonstrating higher incidences of ARF associated with both hypomagnesemia (< 1.7) and hypermagnesemia (> 1.9). After adjusting for potential confounders, both hypomagnesemia (< 1.5 mg/dl) and hypermagnesemia (> 2.3 mg/dl) were associated with an increased risk of developing ARF with odds ratios of 1.69 (95% CI: 1.19-2.36) and 1.40 (95% CI: 1.02-1.91) respectively.

CONCLUSION

Both admission hypomagnesemia and hypermagnesemia were associated with an increased risk for in-hospital ARF.

Renin-angiotensin system inhibitors linked to anemia: a systematic review and meta-analysis

BACKGROUND

The objective of this meta-analysis was to evaluate the risk of anemia in patients who received renin-angiotensin system (RAS) inhibitors.

METHODS

A literature search was performed using MEDLINE, EMBASE, and Cochrane Database of Systematic Reviews from inception through November, 2014. Studies that reported relative risks, odd ratios or hazard ratios comparing the anemia risk in patients who received angiotensin-converting-enzyme inhibitors (ACEIs) or angiotensin II receptor blockers (ARBs) vs. those who did not were included. We performed the prespecified sensitivity analysis including only only studies with confounder adjusted analysis. Pooled risk ratios (RRs) and 95% confidence interval (CI) were calculated using a random-effect, generic inverse variance method.

RESULTS

Seven studies (2 cohort and 5 cross-sectional studies) with 29,061 patients were included in the analysis to assess the risk of anemia and the RAS inhibitors use. The pooled RR of anemia in patients receiving ACEIs was 1.56 (95% CI, 1.40-1.73, I(2) = 17%). When meta-analysis was limited only to studies with confounder adjusted analysis, the pooled RR of anemia in patients using ACEIs was 1.57 (95% CI, 1.43-1.73, I(2) = 0%) The pooled RR of anemia in patients receiving ARBs was 1.60 (95% CI, 1.27-2.00, I(2) = 39%). The meta-analysis of studies with confounder adjusted analysis demonstrated the pooled RR of anemia in patients using ARBs of 1.59 (95% CI, 1.38-1.83, I(2) = 0%).

CONCLUSIONS

Our meta-analysis demonstrates an association between anemia and the use of RAS inhibitors. Hematological parameters should be monitored in patients treated with RAS inhibitors.

High alcohol consumption and the risk of renal damage: a systematic review and meta-analysis

BACKGROUND

The risk of renal damage in patients with high alcohol consumption is controversial. The objective of this meta-analysis was to evaluate the associations between high alcohol consumption and progression of kidney damage including chronic kidney disease (CKD), end-stage renal disease (ESRD) and proteinuria.

METHODS

A literature search was performed using MEDLINE, EMBASE and Cochrane Databases from inception through August 2014 to identify studies investigating the association between high alcohol consumption and CKD, ESRD or proteinuria. Studies that reported odds ratios, relative risks or hazard ratios comparing the risk of CKD, ESRD or proteinuria in patients consuming high amount of alcohol versus those who did not consume alcohol were included. Pooled risk ratios (RRs) and 95% confidence interval (CI) were calculated using a random-effect, generic inverse variance method.

RESULTS

Twenty studies with 292 431 patients were included in our analysis to assess the associations between high alcohol consumption and progression of kidney damage. The pooled RRs of CKD, proteinuria and ESRD in patients with high alcohol consumption were 0.83 (95% CI: 0.71-0.98), 0.85 (95% CI: 0.62-1.17) and 1.00 (95% CI: 0.55-1.82), respectively. Post hoc analysis assessing the sex-specific association between high alcohol consumption and CKD demonstrated pooled RRs of 0.72 (95% CI: 0.57-0.90) in males and 0.78 (95% CI: 0.58-1.03) in females.

CONCLUSIONS

Our study demonstrates an inverse association between high alcohol consumption and risk for developing CKD in males. There is no significant association between high alcohol consumption and the risk for developing proteinuria or ESRD.

The risk of kidney cancer in patients with kidney stones: a systematic review and meta-analysis

BACKGROUND

The objective of this meta-analysis was to evaluate the association between a history of kidney stones and kidney cancer.

METHODS

A literature search was performed from inception until June 2014. Studies that reported odds ratios or hazard ratios comparing the risk of renal cell carcinoma (RCC) and transitional cell carcinoma (TCC) of the upper urinary tract in patients with the history of kidney stones versus those without the history of kidney stones were included. Pooled risk ratios (RRs) and 95% confidence interval (CI) were calculated using a random-effect, generic inverse variance method.

RESULT

Seven studies were included in our analysis to assess the association between a history of kidney stones and RCC. The pooled RR of RCC in patients with kidney stones was 1.76 (95% CI, 1.24-2.49). The subgroup analysis found that the history of kidney stones was associated with increased RCC risk only in males (RR, 1.41 [95% CI, 1.11-1.80]), but not in females (RR, 1.13 [95% CI, 0.86-1.49]). Five studies were selected to assess the association between a history of kidney stones and TCC. The pooled RR of TCC in patients with kidney stones was 2.14 (95% CI, 1.35-3.40).

CONCLUSION

Our study demonstrates a significant increased risk of RCC and TCC in patients with prior kidney stones. However, the increased risk of RCC was noted only in male patients. This finding suggests that a history of kidney stones is associated with kidney cancer and may impact clinical management and cancer surveillance.

Effect of Cystone® on urinary composition and stone formation over a one year period

Kidney stones are a common problem for which inadequate prevention exists. We recruited ten recurrent kidney stone formers with documented calcium oxalate stones into a two phased study to assess safety and effectiveness of Cystone(®), an herbal treatment for prevention of kidney stones. The first phase was a randomized double-blinded 12 week cross over study assessing the effect of Cystone(®) vs. placebo on urinary supersaturation. The second phase was an open label one year study of Cystone(®) to determine if renal stone burden decreased, as assessed by quantitative and subjective assessment of CT. Results revealed no statistically significant effect of Cystone(®) on urinary composition short (6 weeks) or long (52 weeks) term. Average renal stone burden increased rather than decreased on Cystone(®). Therefore, this study does not support the efficacy of Cystone(®) to treat calcium oxalate stone formers. Future studies will be needed to assess effects on stone passage, or on other stone types.

Rituximab treatment of idiopathic membranous nephropathy

Idiopathic membranous nephropathy is a common cause of nephrotic syndrome whose pathogenesis may involve B-cell functions. Rituximab is a monoclonal antibody that binds to the CD20 antigen on B cells thereby deleting them. We conducted an open-label pilot trial of rituximab treatment in 15 severely nephrotic patients with proteinuria refractory to angiotensin-converting enzyme inhibition and/or receptor blockade but with adequately controlled blood pressure. Rituximab was given 2 weeks apart and, at 6 months, patients who remained proteinuric but had recovered B-cell counts were given a second course of treatment. Proteinuria was significantly decreased by about half at 12 months. Of the 14 patients who completed follow-up, full remission was achieved in two and partial remission in six patients based upon the degree of proteinuria. Side effects were minor; however, we found no relationship between the response and number of B cells in the blood, CD20 cells in the kidney biopsy, degree of tubulointerstitial fibrosis, starting proteinuria or creatinine values. Rituximab appears effective in reducing proteinuria in some patients with idiopathic membranous nephropathy but prospective identification of responsive patients was not possible.

Prediction of stone composition from plain radiographs: a prospective study

BACKGROUND AND OBJECTIVES

Stone composition, as reflected in radiographic appearance, is important to help choose between SWL and percutaneous/endoscopic procedures. Predicting a stone's composition accurately from a plain radiograph would be a useful tool in clinical decision-making. However, the ability of physicians to predict composition has not been adequately assessed. A prospective study was designed to quantify the accuracy of a panel of physicians who routinely deal with stones in classifying stone composition solely from radiographs.

MATERIALS AND METHODS

A panel of six members was created to review 100 plain-film radiographs from patients with renal stones of known composition. The panel consisted of two urologists, two radiologists, and two nephrologists, all of whom have expertise in stone disease. If the composition guessed was at least 40% of the total stone composition, the response was deemed correct.

RESULTS

Overall, there was an average 39% correct response score among the six panelists. When the stones were divided by size, 35% were <1 cm, and 65% were larger. The accuracy of chemical composition determination did not improve with greater stone size, nor was there a difference in accuracy for pure and mixed stones. The most frequently misclassified stone was calcium phosphate, with only 14% being correctly diagnosed.

CONCLUSIONS

With a random sampling of plain radiographs, a panel of physicians specializing in stone disease correctly diagnosed the composition of renal calculi less than half of the time without being given clinical information.

Diffuse proliferative glomerulonephritis and acute renal failure associated with acute staphylococcal osteomyelitis

A 72-year-old man developed acute renal failure after coronary bypass surgery that had been complicated by sternal osteomyelitis caused by the Staphylococcus aureus bacterium. Bacteremia and sepsis were not present. Renal biopsy demonstrated a florid, diffuse, proliferative glomerulonephritis with glomerular immune complex deposition. Management included hemodialysis, prolonged antibiotic therapy, and repeated surgical debridement. Spontaneous recovery of renal function occurred after eradication of infection and final surgical wound repair. The relationship between acute bacterial infections and glomerulonephritis and, in particular, the causal role of staphylococcal antigens in the pathogenesis of such lesions is discussed.

The association of nephrolithiasis and autosomal dominant polycystic kidney disease

Despite the frequency and morbidity of nephrolithiasis in autosomal dominant polycystic kidney disease (ADPKD), this association has not been subject to a detailed study. One hundred fifty-one of 751 ADPKD patients seen at the Mayo Clinic between 1976 and 1986 had nephrolithiasis. Seventy-four had passed calculi or had stones surgically removed. Stone analysis was available in 30 patients: uric acid, calcium oxalate, calcium phosphate, and struvite were present in 56.6%, 46.6%, 20%, and 10%, respectively. Calculi were observed in 71 of 79 patients with excretory urograms available for review. Faintly opaque and bull's eye stones, probably containing uric acid, were present in 12.7% and 14.1% of these patients, respectively. Precaliceal tubular ectasia was observed in 15.5%. Ninety-seven patients had preserved renal function (serum creatinine less than 1.5 mg/dL) at the initial evaluation. Six were excluded because they had other known causes of stone disease. The most common metabolic abnormality in the remaining 91 patients was hypocitric aciduria (ten of 15 patients with measurements). The urine pH in the first voided morning specimens (5.66 +/- 0.05) was significantly lower than that of an unselected control population (5.92 +/- 0.03, P less than 0.001). Hyperuricosuria, hyperoxaluria, and hypercalciuria were observed in six of 32 (18.8%), six of 31 (19.4%), and three of 39 (9.7%) patients with preserved renal function. The composition of the stones, the frequency of hypocitric aciduria, and the low urine pH (possibly related to the defect in excretion of ammonia described in ADPKD), suggest that metabolic, along with mechanical, factors are responsible for the frequent occurrence of nephrolithiasis in this disease.

An experimental human model of 1,25-dihydroxyvitamin D-mediated hypercalciuria

An increase in fasting calcium excretion occurs in hypercalciuric patients and has been interpreted by many investigators as evidence for a primary renal tubular leak of calcium. In a recent series of 50 patients with absorptive hypercalciuria, we found a mean increase in fasting fractional calcium excretion (calcium clearance) and provided data suggesting that this leak of calcium was secondary to intestinal hyperabsorption of calcium and suppression of parathyroid secretion. To examine this question, a model of 1,25-dihydroxyvitamin D [1,25-(OH)2D]-mediated hypercalciuria was created by administering a large dose of 1,25-(OH)2D to normal subjects. In addition to the expected features of absorptive hypercalciuria during 1,25-(OH)2D administration, the subjects had an increase in fasting calcium excretion and a marked increase in calcium excretion during a restricted calcium diet, points that might be interpreted as favoring a resorptive and/or renal component to the net hypercalciuria. However, total hydroxyproline excretion remained unchanged (18.0 vs. 19.0 mg/g creatinine), and the increase in fasting calcium excretion was found to reflect an increase in calcium clearance; the latter was inversely correlated with parathyroid function, as determined by fasting measurements of nephrogenous cAMP excretion (r = -0.77; P less than 0.01). We conclude that measurements of calcium excretion in the fasting state or on a restricted calcium diet do not represent valid criteria for differential diagnosis of the hypercalciurias.

Oxalate absorption and postprandial urine supersaturation in an experimental human model of absorptive hypercalciuria

The effect of 1.25-dihydroxyvitamin D [1,25-(OH)2D] on dietary oxalate absorption and postprandial urine supersaturation with calcium oxalate was determined in 11 normal subjects. 1,25-(OH)2D increased the urinary excretion of orally administered [14C]oxalate in the 8 h period after a liquid meal containing 1.875 mmol of calcium and 0.83 mmol of oxalate (P less than 0.01), and during a 48 h period when the subjects ingested a diet containing 25 mmol of calcium and 3.3 mmol of oxalate/day (P less than 0.01); however, 1,25-(OH)2D administration had no effect on [14C]oxalate excretion when calcium was removed from the liquid meal. 1.25-(OH)2D increased 24 h urinary oxalate excretion from 28.7 +/- 2.1 mmol/mol of creatinine to 36.8 +/- 2.6 mmol/mol of creatinine (P less than 0.05) on the 10 mmol/day calcium diet and from 26.4 +/- 2.9 to 33.2 +/- 2.2 mmol/mol of creatinine (P less than 0.1) on the 25 mmol/day calcium diet. A linear correlation (r = 0.72) was found between plasma 1,25-(OH)2D levels and urinary [14C]oxalate excretion after the liquid meal. 1,25-(OH)2D administration produced postprandial supersaturation of urine with calcium oxalate and calcium oxalate crystalluria. These studies suggest that 1,25-(OH)2D increases oxalate absorption (and urinary excretion) by increasing calcium absorption, which results in less binding of calcium to oxalate in the intestine; therefore more oxalate is available for absorption. The combined effect of increased calcium and oxalate absorption results in postprandial supersaturation of urine with calcium oxalate, with resultant crystalluria.

When should the stone patient be evaluated? Limited evaluation of single stone formers

The purpose of a metabolic stone evaluation is to provide information that will influence treatment. For stone formers who have not tried a conservative diet-fluid program, only a limited metabolic evaluation is indicated. An extensive metabolic evaluation is reserved for stone formers who actively grow calculi despite conservative treatment.

The stone clinic effect in patients with idiopathic calcium urolithiasis

The "stone clinic effect" refers to the effect of encouraging a high intake of fluid and avoiding dietary excesses on stone formation and growth in patients with urolithiasis. To determine the extent of this effect we reviewed the clinical courses of 108 patients with idiopathic calcium urolithiasis and indeterminant metabolic activity. There was no evidence of stone growth or new stone formation (metabolic inactivity) after a mean followup of 62.6 months in 63 of the 108 patients (58.3 per cent), including 12 of 17 (70.6 per cent) with hypercalciuria and 7 of 15 (46.7 per cent) with hyperuricosuria. Comparison of initial and followup 24-hour urine volumes demonstrated a significant increase in patients who were metabolically inactive at followup (p less than 0.0005), while no increase was detected in patients who were metabolically active at followup. We recommend that specific drug therapy should not be given to patients with idiopathic calcium urolithiasis until the stone clinic effect has been evaluated.

Use of combined plasmapheresis and immunosuppression in the treatment of Goodpasture's syndrome

Five consecutive patients with well-documented Goodpasture's syndrome were treated with plasmapheresis and immunosuppression. In all patients, the antiglomerular basement-membrane antibody titers decreased with treatment. In three patients, hemoptysis responded promptly to plasmapheresis. Two patients presenting with severe renal failure required chronic dialysis, and three patients who had serum creatinine levels less than 2.1 mg/dl before treatment improved or had stabilization of their renal function. We confirm that the use of plasmapheresis and immunosuppression is a promising method of treatment in some patients with Goodpasture's syndrome.