Dear Editor,
We have previously reported unusually high rates of elevated glycated hemoglobin (HbA1c) levels in patients with COVID-19 admitted to an ICU between March 11 and April 29, 2020 [1]. Since then, our Tyrolean multicenter COVID-19 Intensive Care Unit Registry (Tyrol-CoV-ICU-Reg) [2] has considerably increased, surpassing 500 patients. To re-evaluate our previously reported findings, we included 306 additional patients in this analysis, who were admitted between April 30, 2020 and May 31, 2021, for whom an admission HbA1c was available. Details of our registry have been reported before [2].
Admission HbA1c was now available in 350 patients. We were able to confirm our finding that HbA1c was elevated (i.e., HbA1c ≥ 5.7%) in most patients (85.1%). However, only 31.7% had a history of diabetes mellitus (DM) or prediabetes (Table 1). Median HbA1c at admission was significantly higher in patients with HbA1c ≥ 6.5% and a history of DM than without history of DM (Mann–Whitney U p < 0.001). Furthermore, a weak correlation for HbA1c and BMI could be established (Pearson R = 0.27, p < 0.001).
Table 1
General characteristics of included patients stratified by history of diabetes mellitus/prediabetes and HbA1c
Overall | No history of diabetes mellitus/prediabetes | History of diabetes mellitus/prediabetes | |||||
|---|---|---|---|---|---|---|---|
HbA1c < 5.7% | HbA1c 5.7 < 6.5% | HbA1c ≥ 6.5% | HbA1c < 5.7% | HbA1c 5.7 < 6.5% | HbA1c ≥ 6.5% | ||
n | 350 | 49 | 138 | 52 | 3 | 18 | 90 |
Age [years] (median [IQR]) | 68.00 [58.00, 76.00] | 67.00 [54.00, 77.00] | 70.00 [61.00, 77.00] | 65.00 [55.50, 75.25] | 63.00 [56.50, 68.00] | 72.00 [62.50, 74.75] | 68.00 [58.25, 76.75] |
Male—no. (%) | 250 (71.4) | 32 (65.3) | 102 (73.9) | 40 (76.9) | 3 (100.0) | 11 (61.1) | 62 (68.9) |
BMI [kg/m2] (median [IQR]) | 27.76 [25.14, 31.22] | 25.12 [22.10, 27.53] | 27.47 [24.97, 30.66] | 29.45 [26.41, 31.51] | 25.13 [24.18, 29.52] | 28.73 [26.93, 31.71] | 29.39 [26.51, 32.92] |
HbA1c [%] (median [IQR]) | 6.30 [5.90, 6.80] | 5.40 [5.10, 5.50] | 6.10 [5.90, 6.20] | 6.70 [6.60, 7.12] | 5.50 [5.50, 5.55] | 6.10 [5.93, 6.30] | 7.60 [6.80, 8.88] |
IMV—no. (%) | 194 (55.4) | 28 (57.1) | 80 (58.0) | 31 (59.6) | 1 (33.3) | 9 (50.0) | 45 (50.0) |
AKI—no. (%) | |||||||
no AKI | 233 (67.0) | 33 (68.8) | 94 (68.6) | 31 (59.6) | 2 (66.7) | 12 (66.7) | 61 (67.8) |
KDIGO I | 38 (10.9) | 4 (8.3) | 16 (11.7) | 7 (13.5) | 0 (0.0) | 2 (11.1) | 9 (10.0) |
KDIGO II | 18 (5.2) | 2 (4.2) | 7 (5.1) | 2 (3.8) | 0 (0.0) | 1 (5.6) | 6 (6.7) |
KDIGO III | 59 (17.0) | 9 (18.8) | 20 (14.6) | 12 (23.1) | 1 (33.3) | 3 (16.7) | 14 (15.6) |
RRT—no. (%) | 55 (15.7) | 9 (18.4) | 18 (13.0) | 10 (19.2) | 1 (33.3) | 5 (27.8) | 12 (13.3) |
vv-ECMO | 18 (5.1) | 4 (8.2) | 5 (3.6) | 6 (11.5) | 0 (0.0) | 0 (0.0) | 3 (3.3) |
IMV [days] (median [IQR]) | 14.0 [8.0, 24.3] | 10.5 [5.5, 20.8] | 13.0 [8.0, 25.3] | 15.00 [12.5, 26.50] | 7.0 [7.0, 7.0] | 16.0 [5.0, 29.0] | 13.0 [8.0, 24.0] |
RRT [days] (median [IQR]) | 11.0 [3.0, 25.5] | 12.0 [6.0, 13.0] | 16.5 [3.5, 27.5] | 13.50 [9.5, 25.8] | 7.0 [7.0, 7.0] | 1.0 [1.0, 4.0] | 4.0 [1.8, 14.5] |
ECMO [days] (median [IQR]) | 23.5 [12.5, 28.8] | 15.0 [14.0, 16.5] | 26.0 [14.0, 26.0] | 28.50 [16.0, 38.0] | NA [NA, NA] | NA [NA, NA] | 27.0 [19.5, 28.5] |
SAPS III score (median [IQR]) | 55.0 [48.0, 63.0] | 59.0 [49.5, 69.8] | 54.5 [49.0, 63.3] | 52.50 [47.0, 62.5] | 53.0 [49.5, 61.5] | 55.5 [50.8, 59.3] | 55.0 [48.0, 62.0] |
Hospital LOS (median [IQR]) | 23.0 [14.0, 39.5] | 26.5 [14.0, 40.0] | 22.0 [15.0, 35.0] | 24.00 [12.0, 45.5] | 44.0 [33.5, 61.5] | 30.5 [14.8, 45.5] | 21.0 [13.0, 36.5] |
ICU LOS (median [IQR]) | 11.0 [5.0, 23.0] | 10.0 [4.0, 23.0] | 13.0 [7.0, 23.0] | 16.50 [5.0, 29.0] | 5.0 [3.5, 10.5] | 9.5 [6.0, 30.5] | 10.0 [5.0, 21.0] |
Known comorbidity*—no. (%) | |||||||
Cardiovascular | 142 (40.6) | 21 (42.9) | 56 (40.6) | 18 (34.6) | 2 (66.7) | 12 (66.7) | 33 (36.7) |
Arterial hypertension | 221 (63.1) | 26 (53.1) | 82 (59.4) | 25 (48.1) | 3 (100.0) | 15 (83.3) | 70 (77.8) |
Renal | 78 (22.3) | 12 (24.5) | 24 (17.4) | 10 (19.2) | 2 (66.7) | 8 (44.4) | 22 (24.4) |
Liver | 28 (8.0) | 3 (6.1) | 12 (8.7) | 6 (11.5) | 0 (0.0) | 1 (5.6) | 6 (6.7) |
Metastatic disease | 2 (0.6) | 0 (0.0) | 0 (0.0) | 1 (1.9) | 0 (0.0) | 0 (0.0) | 1 (1.1) |
Hematological malignancy | 18 (5.1) | 3 (6.1) | 7 (5.1) | 2 (3.8) | 0 (0.0) | 0 (0.0) | 6 (6.7) |
Non-hematological malignancy | 28 (8.0) | 3 (6.1) | 15 (10.9) | 3 (5.8) | 0 (0.0) | 0 (0.0) | 7 (7.9) |
COPD | 17 (4.9) | 4 (8.2) | 8 (5.8) | 0 (0.0) | 0 (0.0) | 1 (5.6) | 4 (4.4) |
Asthma | 50 (14.3) | 4 (8.2) | 18 (13.0) | 8 (15.4) | 1 (33.3) | 2 (11.1) | 17 (18.9) |
Respiratory disease—others | 27 (7.7) | 2 (4.1) | 12 (8.7) | 3 (5.8) | 0 (0.0) | 2
(11.1) | 8 (8.9) |
Diabetes mellitus—no. (%) | |||||||
No pre-known dysglycemia | 239 (68.3) | 49 (100.0) | 138 (100.0) | 52 (100.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
Prediabetes | 12 (3.4) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 4 (22.2) | 8 (8.9) |
DM Type I | 4 (1.1) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 2 (11.1) | 2 (2.2) |
DM Type II | 94 (26.9) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 3 (100.0) | 11 (61.1) | 80 (88.9) |
DM (other Type, e.g., MODY) | 1 (0.3) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 1 (5.6) | 0 (0.0) |
ICU mortality—no. (%) | 90 (25.7) | 13 (26.5) | 35 (25.4) | 14 (26.9) | 0 (0.0) | 3 (16.7) | 25 (27.8) |
Hospital mortality—no. (%) | 100 (28.6) | 15 (30.6) | 37 (26.8) | 15 (28.8) | 0 (0.0) | 3 (16.7) | 30 (33.3) |
Anzeige
There was a trend toward longer duration of invasive mechanical ventilation (IMV) in patients with higher HbA1c, albeit non-significant. In patients with history of DM, a tendency toward increased mortality associated with elevated HbA1c was observed, but the groups differed considerably in size (Table 1). When comparing patients with an HbA1c ≥ 6.5% requiring IMV to the rest of our cohort, ICU (40.8% vs. 21.5%; Χ2 p = 0.001) and hospital mortality (42.1% vs. 24.8%; Χ2 p = 0.005) were significantly increased.
In a multivariate logistic regression model including HbA1c, history of DM and IMV, odds ratios (ORs) for hospital death were higher for patients with elevated HbA1c ≥ 6.5%. However, this effect seems to be mainly driven by IMV. A significant association with hospital mortality was shown for treatment with IMV, age and SAPS III score (Table 2).
Table 2
Logistic regression analysis for prediction of hospital mortality
Univariate analysis | Multivariate analysis | |||
|---|---|---|---|---|
OR (95% CI) | p | OR (95% CI) | p | |
Age | 1.08 (1.05–1.11) | < 0.001 | 1.07 (1.04–1.11) | < 0.001 |
Sex (male) | 1.29 (0.77–2.21) | 0.350 | ||
BMI | 1.00 (0.97–1.05) | 0.737 | ||
Number of comorbidities | 1.29 (1.13–1.49) | < 0.001 | 1.24 (1.04–1.48) | 0.018 |
SAPS III score | 1.08 (1.05–1.11) | < 0.001 | 1.05 (1.02–1.08) | 0.001 |
IMV | 3.08 (1.87–5.21) | < 0.001 | ||
Ref: HbA1c < 6.5 and no history of DM | ||||
HbA1c < 6.5 and history of DM | 0.43 (0.20–1.35) | 0.194 | ||
HbA1c ≥ 6.5 and no history of DM | 1.05 (0.52–2.05) | 0.883 | ||
HbA1c ≥ 6.5 and history of DM | 1.30 (0.75–2.23) | 0.346 | ||
Ref: HbA1c < 6.5 and no history of DM and no IMV | ||||
HbA1c < 6.5 and history of DM, no IMV | 0.30 (0.01–2.16) | 0.304 | ||
HbA1c ≥ 6.5 and no history of DM, no IMV | 0.31 (0.02–1.99) | 0.298 | ||
HbA1c ≥ 6.5 and history of DM, no IMV | 2.18 (0.71–6.71) | 0.168 | ||
HbA1c < 6.5 and no history of DM, IMV | 3.94 (1.72–9.66) | 0.002 | ||
HbA1c < 6.5 and history of DM, IMV | 0.78 (0.04–6.16) | 0.832 | ||
HbA1c ≥ 6.5 and no history of DM, IMV | 6.36 (2.02–20.85) | 0.002 | ||
HbA1c ≥ 6.5 and history of DM, IMV | 3.66 (1.38–10.17) | 0.011 | ||
In this follow-up analysis, including 306 additional cases, we were able to confirm our previous findings [1] of an extremely high incidence of elevated HbA1c in critically ill COVID-19 patients. While other pre-admission comorbidities (e.g., arterial hypertension) were more common than DM, chronic dysglycemia as defined by an HbA1c ≥ 5.7% was the predominant factor in our cohort of critically ill COVID-19 patients. We still did not find a strong association between elevated HbA1c and ICU/hospital mortality, but in the subgroup of patients requiring mechanical ventilation, mortality was significantly increased in patients with HbA1c ≥ 6.5%. This is in line with other cohorts of COVID-19 patients, showing that admission blood glucose may be a more relevant predictor for mortality than HbA1c [3]. Admission blood glucose may be interpreted as a biomarker of systemic inflammation on admission, whereas HbA1c represents a marker of glucose control of the past three months [4, 5]. Unfortunately, blood glucose measurements during ICU admission were not recorded in our registry. Furthermore, follow-up of reported patients would be necessary to confirm diagnosis of previously unrecognized diabetes mellitus by admission HbA1c. We also have to acknowledge a potential selection bias for patients in whom HbA1c was measured. BMI has recently been reported as an important risk factor for severe COVID-19 [3]. Though we could establish a weak correlation between BMI and HbA1c, we want to emphasize that prediabetes/and diabetes may exist independently and thus remain an independent risk factor for ICU admission in COVID-19. A possible explanation may be a pathological inflammatory response in diabetic patients [6]. When exposed to an additional inflammatory stimulus, such as mechanical ventilation on top of COVID-19, outcome may be impaired. This finding warrants further research in terms of risk stratification at ICU admission.
Acknowledgements
Collaborators and their Institutions: Bellmann Romuald, Andreas Peer, Ditlbacher Adelheid, Haßlacher Julia, Lehner Georg F. and Perschinka Fabian: Division of Intensive Care and Emergency Medicine, Department of Internal Medicine, Medical University Innsbruck (Austria). Hasibeder Walter: Department of Anesthesiology and Critical Care Medicine, Hospital St. Vinzenz Zams (Austria). Krismer Christoph: Department of Internal Medicine, Hospital St. Vinzenz Zams (Austria). Pechlaner Agnes: Department of Internal Medicine, Hospital St. Vinzenz Zams (Austria). Eschertzhuber Stephan: Department of Anesthesia and Intensive Care Medicine, Hospital Hall (Austria). Zagitzer-Hofer Stefanie: Department of Anesthesia and Intensive Care Medicine, Hospital Hall (Austria). Foidl Eva: Department of Anesthesia and Intensive Care Medicine, Hospital Kufstein (Austria). Weilguni Isabella: Department of Anesthesia and Intensive Care Medicine, Hospital Kufstein (Austria). Haslauer-Mariacher Stefanie: Department of Anesthesia and Intensive Care Medicine, Hospital Kufstein (Austria). Kalenka Armin: Department of Anesthesia and Intensive Care Medicine, Hospital Kufstein (Austria). Ribitsch Alexandra: Department of Internal Medicine, Hospital Lienz (Austria). Mayr Andreas: Department of Anesthesia and Intensive Care Medicine, Hospital Lienz (Austria). Ladner Eugen: Department of Anesthesia and Intensive Care Medicine, Hospital Reutte (Austria), Heiner Tatjana: Department of Anesthesia and Intensive Care Medicine, Hospital Reutte (Austria). Mayr-Hueber Bernhard: Department of Anesthesia and Critical Care Medicine, Hospital Schwaz (Austria). Stögermüller Birgit: Department of Anesthesia and Critical Care Medicine, Hospital Schwaz (Austria). Kirchmair Lukas: Department of Anesthesia and Critical Care Medicine, Hospital Schwaz, (Austria). Reitter Bruno: Department of Anesthesia and Intensive Care Medicine, Hospital St. Johann in Tyrol (Austria). Potocnik Miriam: Department of Anesthesia and Intensive Care Medicine, Hospital St. Johann in Tyrol (Austria). Mathis Simon: Department of Anesthesia and Critical Care Medicine, Medical University Innsbruck (Austria). Fiala Anna: Department of Anesthesia and Critical Care Medicine, Medical University Innsbruck (Austria). Brunner Jürgen: Department of Pediatrics, Medical University Innsbruck, (Austria). Claudius Thomé: Department of Neurosurgery, Medical University Innsbruck (Austria).
Anzeige
Declarations
Ethics approval and consent to participate
This study was approved by the ethics committee of the Medical University Innsbruck (# 1099/2020). Informed consent or post hoc informed consent was obtained.
Consent for publication
Not applicable—manuscript contains no individual patient data.
Competing interests
The authors declare that they have no competing interests.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.