Early Identification Of Patients Who Will Meet 24-hour Fluid Output Threshold For Chest Tube Removal After Lung Resection
Jason L. Azzi, Bram Gottlieb, Donna E. Maziak, Andrew J.E. Seely, Farid M. Shamji, Sudhir Sundaresan, Patrick J. Villeneuve, Sebastien Gilbert.
University of Ottawa, Ottawa, ON, Canada.
BACKGROUND - Most chest tube (CT) removal protocols include a fluid output threshold based on 24-hour observation periods. This study aims to evaluate if fluid output measured over a shorter period can reliably predict if the 24-hour fluid output will be at or below target for removal. METHODS - Following lung resection, pleural fluid output data prospectively recorded by a digital drainage system, was smoothed using Locally Weighted Scatterplot Smoothing (LOWESS) and retrospectively analyzed. Different 24-hour fluid output thresholds for CT removal were evaluated (400mL, 250mL, and 20% of whole-body lymphatic flow (WBLF)). Every available 6-, 8-, and 12-hour fluid output measurement was compared to prorated 24-hour output thresholds. The 6-, 8-, and 12-hour fluid outputs were used to predict if the 24-hour fluid output threshold would be met 18, 16, and 12 hours later, respectively. McNemar’s test was used to assess statistical significance. RESULTS - From September 2015 to June 2017, digitally recorded pleural fluid output data was available in 120 patients who underwent lung resections. The median age was 64.7±11.8 years and there were more females (61%; 73/120). Most patients had either a lobectomy (72%; 86/120) or a wedge resection (12%; 14/120) for malignant disease (89%; 107/120). The most common approach was VATS (85%; 102/120). Depending on the time interval and the 24-hour threshold used, specificity, positive predictive value (PPV) and false positive rate (FPR) ranged from 84-94%, 76-92% and 6-16%, respectively. (Table 1) FPR decreased with increasing duration of interval fluid output measurements. There was a statistically significant difference in performance between specificity and PPV between 6-, 8- and 12-hour outputs when a 250 mL or WBLF 24-hour output threshold for removal was used, except between a 6-hour and 8-hour output when the WBLF 24-hour output threshold was used. CONCLUSIONS - Pleural fluid drainage data at 6-, 8- and 12-hour intervals accurately identifies patients who will meet 24-hour fluid output threshold for CT removal. There may be an opportunity to gain clinical efficiency without compromising safety by using shorter fluid drainage periods to determine if chest tubes can be removed sooner based on 24-hour fluid output criteria.
| 24-hour threshold | Time interval (hours) | Specificity (95% CI) | Positive predictive value (95% CI) | False positive rate (95% CI) |
| 20% Whole body lymphatic flow | 6 | 84% (82-86) | 84% (82-85) | 16% (14-18) |
| 20% Whole body lymphatic flow | 8 | 87% (85-88) | 85% (83-87) | 13% (12-15) |
| 20% Whole body lymphatic flow | 12 | 91% (90-93) | 87% (86-89) | 9% (7-10) |
| 400 mL | 6 | 86% (84-88) | 91% (89-92) | 14% (12-16) |
| 400 mL | 8 | 87% (85-89) | 91% (90-92) | 13% (11-15) |
| 400 mL | 12 | 91% (90-93) | 92% (90-93) | 9% (7-10) |
| 250 mL | 6 | 88% (87-89) | 76% (74-78) | 12% (11-13) |
| 250 mL | 8 | 90% (89-91) | 78% (76-80) | 10% (9-11) |
| 250 mL | 12 | 94% (93-95) | 82% (79-84) | 6% (5-7) |
Back to 2018 Thoracic Track