Accuracy of Lower Temperature Thresholds in Detecting COVID-19 in Hemodialysis Patients
Bibliographic record
Abstract
Background: Patients receiving in-center hemodialysis (HD) are uniquely vulnerable to COVID-19 yet identifying infected individuals may be challenging. They may not present with typical symptoms and low basal body temperature may impair detection of fever. We studied the accuracy of temperature thresholds in detecting COVID-19 in HD patients. Methods: We retrospectively studied all patients between March 24-May 14, 2020 from a single HD unit (Hôpital du Sacré-Coeur) in Montreal, Canada, where COVID-19 is highly prevalent. All patients who presented with symptoms or contact exposure were tested by nasopharyngeal swab. Prompted by an outbreak, systematic testing of all HD patients was started on April 18th. Basal temperature was defined as the average predialysis temperature from weeks -1 to -3 before testing. Diagnostic performance was determined for various temperature thresholds defined a priori. Results: Of 205 in-center HD patients, 34 developed COVID-19 during the study period. Of these, 21 (61%) were hospitalised, 4 (11%) required intensive care and 9 (26%) died. Baseline characteristics are presented in Table 1. Less than a third had typical symptoms. Thresholds of ≥ 37.3 °C and “basal temperature +0.5 °C” had similar moderate sensitivity and high specificity in predicting COVID-19 (Table 2). Combining symptoms and either of these thresholds improved sensitivity to 85%. - Characteristics COVID-19 negative (n=171) COVID-19 positive (n=34) p-value Male sex 59% 56% 0.7 Age 71 (60, 81) 76 (68, 85) 0.03 Black race 20% 38% 0.02 Living in long-term care facilities 11% 32% 0.001 Primary kidney disease 0.5 Diabetic 47% 32% Hypertensive 22% 47% Glomerulonephritis 14% 9% Diabetes 54% 47% 0.4 Ischemic heart disease 37% 32% 0.6 Heart failure 9% 24% 0.01 Indication for screening <0.001 Symptoms without fever 2% 24% Fever (at home, or ≤37.5 in HD) 1% 47% Contact with positive case 4% 9% Systematic screening 94% 21% Temperature at screening (°C) 36.5 (36.3, 37.0) 37.5 (37.0, 38.0) <0.001 Basal temperature (°C) 36.5 (36.3, 36.7) 36.6 (36.4, 36.9) 0.1 Change from basal (°C) 0.1 (-0.3, 0.5) 0.9 (0.3, 1.5) <0.001 Above 37°C 28% 77% <0.001 Above 37.5 °C 8% 65% <0.001 Above 38 °C 0% 27% <0.001 Above basal 56% 87% 0.001 Above hasal + 0.5 °C 25% 70% <0.001 Above hasal + 1°C 1% 43% <0.001 Presented as Percent and median (25th-75th percentiles) Conclusions: Less than one third of HD patients have typical symptoms of COVID-19 or fever >38.0°C. Pre-dialysis temperature >37.3°C or 0.5°C above basal temperature markedly improves sensitivity for detection of COVID-19 in asymptomatic HD patients. A screening strategy combining symptom questionnaires and pre-dialysis temperature monitoring should be used in HD units in regions of high COVID-19 prevalence. Table 2: - Diagnostic performance of various thresholds Pre-dialysis temperature Sensitivity Specificity Positive predictive value Negative predictive value Positive likelihood ratio Negative likelihood ratio Above 37 °C 77% 72% 35% 94% 2.7 0.33 Above 37.3 °C 65% 92% 61% 93% 7.9 0.38 Above 37.5 °C 59% 96% 74% 92% 14.3 0.43 Above 38 °C 27% 100% 100% 87% ∞ 0 Above basal + 0.5 °C 70% 93% 33% 93% 10.7 0.32 Above basal + 1 °C 43% 91% 93% 91% 4.7 0.62
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.007 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".