Ocular involvement in STEC-associated hemolytic uremic syndrome.

BACKGROUND: Hemolytic uremic syndrome (HUS) is a systemic thrombotic microangiopathy characterized by hemolytic anemia, thrombocytopenia, and variable kidney involvement. Extrarenal thrombotic microangiopathy occurs in central nervous system (CNS), colon, and other organ systems, but ocular involvement is rarely recognized. This study aimed to analyze frequency and severity of ocular involvement in STEC-HUS, and the relationship between ocular involvement and disease severity, with emphasis on CNS, kidney, and colonic disease. METHODS: Prospective, longitudinal, observational study. INCLUSION CRITERIA: STEC-HUS patients September 2014-January 2019. Funduscopic examination (FE) was performed within 48 h of admission. We evaluated severity of CNS disease, kidney involvement, and presence of hemorrhagic colitis (HC). RESULTS: Ninety-nine patients were included (female 52), mean age 39.4 months (DE: 29.8; range 9-132). Thirteen patients (13.1%) had abnormal FE, 10 showing variable degrees of hemorrhagic exudates and 2 with typical Purtscher-like retinopathy. Other findings included tortuous vascularity, cotton wool spots, and transient retinal edema. CNS involvement was present in 16/99 patients, severe in 12 (75%). Abnormal FE occurred in 5/12 (31%) patients with severe CNS involvement vs. 8/87 (9.2%) with mild, moderate, or no CNS disease (p = 0.0191). Abnormal FE was present in 2/33 (6%) patients without dialysis vs. 11/66 (16.6%) requiring dialysis (p = 0.20). Finally, there were FE abnormalities in 6/20 patients with HC vs. 7/79 without HC (p = 0.012). CONCLUSIONS: FE abnormalities were present in 13% of HUS patients. Abnormal FE significantly associated with more severe disease, including severe CNS involvement and HC. We suggest FE should be performed in severe HUS, especially in cases with severe CNS disease. A higher resolution version of the Graphical abstract is available as Supplementary information.

Hypoalbuminemia: a risk factor in patients with STEC-associated hemolytic uremic syndrome.

BACKGROUND: We aimed to determine the prevalence of hypoalbuminemia in STEC-HUS patients with hemorrhagic colitis (HC) and whether serum albumin level (SAL), leukocyte count, hematocrit and serum sodium level (SSL) are prognostic markers of HC, central nervous system disease (CNSd) and/or dialysis requirement and evaluate if hypoalbuminemia is associated with fecal protein losses. METHODS: We prospectively evaluated STEC-HUS patients treated at our institution from 9/2011 to 2/2019, analyzing the presence of HC, CNSd and dialysis requirement and SAL, SSL, leukocytes, hematocrit and α1-antitrypsin clearance. RESULTS: We evaluated 98 patients, with mean age of 33.3 months. SAL ≤ 29.5 g/l, > 24,600 leukocytes/mm3 and hematocrit > 30% behave as independent prognostic markers for HC. SAL ≤ 28 g/l, > 25,200 leukocytes/mm3 and hematocrit > 30% behave as prognostic markers for CNSd. SAL ≤ 31.6 g/l, > 13,800 leukocytes/mm3, hematocrit > 18.9% and hyponatremia (≤ 132 mEq/l) behave as prognostic markers for dialysis requirement. However, in multivariate logistic regression models, only hypoalbuminemia behaved as a risk factor for HC, CNSd and dialysis. α1-antitrypsin clearance was performed in 69 patients and was high in 9/69 (13%), only 4 with HC. No significant association was observed between α1-antitrypsin clearance and albuminemia (χ2 = 0.1076, p = 0.7429) as well as α1-antitrypsin clearance and HC (χ2 = 1.7892, p = 0.1810). CONCLUSIONS: Almost all patients with HC had hypoalbuminemia, which behaves as a risk factor for HC, CNSd and dialysis requirement. No significant association was observed between elevated α1-antitrypsin clearance and hypoalbuminemia nor between elevated α1-antitrypsin clearance and HC. These findings could be related to the small number of evaluated patients.

Relationship between red blood cell transfusion requirements and severity of renal disease during the acute stage of hemolytic uremic syndrome.

BACKGROUND: We performed a retrospective evaluation of patients with diarrhea-associated hemolytic uremic syndrome (D + HUS) with the aims of: (1) determining the rate of red blood cell (RBC) transfusions; (2) establishing the relationship between need for RBC transfusion and severity of renal involvement; (3) determining whether precise measurements of lactic dehydrogenase (LDH) levels can predict the rate of hemolysis and severity of renal disease. METHODS: A total of 288 patients with D + HUS were retrospectively divided into three groups based on dialysis treatment: group 1, no dialysis treatment (144 patients); group 2, dialysis for 1-10 days (67 patients); group 3, dialysis for ≥11 days (77 patients). RESULTS: Of the patients in groups 1, 2 and 3, 73.6, 86.5 and 83.1%, respectively, required at least one RBC transfusion. The number of RBC transfusions in groups 1, 2 and 3 was 163, 107 and 162, respectively. Comparison of the groups revealed that the number of RBC transfusions was significantly higher in patients in groups 2 and 3 than in those in group 1 (p = 0.0001). Most RBC transfusions (94.2%) occurred during the first 2 weeks of the disease. The median peak LDH level was 2091 U/l in 32 patients with no RBC transfusion (group A), 3900 U/l in 73 patients with one transfusion (group B) and 6378 U/l in 62 patients with two or more transfusions (group C). Patients who received two or more RBC transfusions had a significantly higher median peak LDH level than those who did not receive RBC transfusions or received only one transfusion. This difference was also observed between patients who received only one RBC transfusion and those who did not receive any transfusions (p < 0.00001). Comparison of LDH levels on admission and peak LDH levels among patients in groups A, B and C revealed that 28/32 patients in group A, 56/73 patients in group B and 33/62 patients in group C had a stable LDH level, suggesting that patients with a stable LDH level require fewer RBC transfusions (p ≤ 0.006). Finally, we evaluated the possibility of an association between peak LDH levels and the degree of renal disease. The median peak LDH level in patients of group 1, 2 and 3 was 3538 (range 756-9373), 5165 (451-9205) and 7510 (1,145-16,340) U/l, respectively. Patients with >10 days of dialysis (group 3) had the highest LDH levels, followed by patients with 1-10 days of dialysis (group 2) and then by patients with no dialysis requirements (group 1) (p < 0.00001). CONCLUSIONS: The rate of RBC transfusion was higher in patients with the most severe renal injury, and most were performed during the first 2 weeks of the disease. Patients with stable LDH levels seemed to require fewer RBC transfusions. Median peak LDH levels were significantly higher in the group of patients with the most severe renal disease.

Hemorrhagic colitis in postdiarrheal hemolytic uremic syndrome: retrospective analysis of 54 children.

Hemorrhagic colitis (HC) is a severe manifestation of the hemolytic uremic syndrome (HUS). We performed a retrospective analysis of patients with HC with the following aims: (1) to characterize the clinicopathologic features; (2) to evaluate mortality rate; (3) to analyze severity of renal and central nervous system (CNS) disease. Patients with HC assisted between 1981-2009 were evaluated and compared with a control group of 137 patients without HC. Among 987 patients with diarrheal prodrome (D) + HUS, 54 (5.5%) presented HC. Clinical findings included abdominal pain (96%), distension (93%), hematochezia (44%), and abdominal mass (11%). Surgery was indicated in 35 patients (65%), and 17 (48.5%) required bowel resection. Transverse and ascending colon were most frequently affected. Macroscopic evaluation showed bowel necrosis (18) and perforation (12). Histologic evaluation (29) showed that 25 (86.2%) had necrosis of the affected segment (transmural in 21). A leukocyte count >20,000/mm(3) and hematocrit >30% were more common in HC patients than in controls (p < 0.001 and p < 0.0001, respectively). Mortality rate was higher in HC patients (33.3%) than in controls (1.4%; p < 0.0001). Dialysis >10 days, seizures, and coma were more frequent in HC patients than in controls (p < 0.0001). In summary, most patients had prominent abdominal findings, and almost 2/3 patients required surgery. Transverse/ascending colon was most affected, and the main histologic finding was transmural necrosis. Higher hematocrit and leukocytosis were frequent. Mortality rate was extremely high, and most had long-lasting anuria and severe neurologic involvement.

Long-term follow-up of Argentinean patients with hemolytic uremic syndrome who had not undergone dialysis.

We examined the records of patients with hemolytic uremic syndrome, who had not undergone dialysis during the acute stage, with the aims of evaluating: (1) the outcome after at least 5 years of follow-up; (2) the value of peak serum creatinine as a prognostic marker; (3) the relationship between outcome and time to normalization of renal function. From 1968 to 2000, 1,179 patients were assisted. Forty-two patients (3.6%) died during the acute stage, 478 patients (40.5%) required dialysis and 659 patients (55.9%) did not undergo dialysis; 529 non-dialysis patients were lost to follow-up. The remaining 130 patients were classified into four groups: group I, complete recovery; group II, with two subgroups, IIa, microalbuminuria, and IIb, proteinuria and/or high blood pressure, both with normal renal function; group III, chronic renal failure; and group IV, end-stage renal disease. We analyzed the relationship between final outcome and: (1) peak creatinine (the highest of at least two determinations) during the acute stage and (2) time to normalization of urea and/or creatinine after the acute stage. After a mean follow-up time of 147.1 months (range 60-362 months), group I had 83 patients (63.9%), group IIa had 27 (20.8%), group IIb had 15 (11.5%) and group III had 5 (3.8%). The value of peak serum creatinine concentration was available for 57 patients. On the last clinical visit, eight out of 26 (30.7%) patients with peak serum creatinine equal to or higher than 1.5 mg/dl were in groups IIb and III versus one out of 31 (3.2%) patients with lower values (P < or = 0.007). Finally, six out of 28 patients (21%) whose renal function had normalized after 15 days from diagnosis were in groups IIb-III versus 8/82 (9.7%) whose renal function had normalized within 15 days (P = 0.18). After a mean period of follow-up of 12 years, 15% of a selected patient group had developed proteinuria, high blood pressure or chronic renal failure, and 21% had developed microalbuminuria. Peak serum creatinine during the acute stage was useful as a prognostic indicator. Patients whose renal function required more time to normalize did not have a worse outcome.