Causal inference using observational intensive care unit data: a scoping review and recommendations for future practice.

This scoping review focuses on the essential role of models for causal inference in shaping actionable artificial intelligence (AI) designed to aid clinicians in decision-making. The objective was to identify and evaluate the reporting quality of studies introducing models for causal inference in intensive care units (ICUs), and to provide recommendations to improve the future landscape of research practices in this domain. To achieve this, we searched various databases including Embase, MEDLINE ALL, Web of Science Core Collection, Google Scholar, medRxiv, bioRxiv, arXiv, and the ACM Digital Library. Studies involving models for causal inference addressing time-varying treatments in the adult ICU were reviewed. Data extraction encompassed the study settings and methodologies applied. Furthermore, we assessed reporting quality of target trial components (i.e., eligibility criteria, treatment strategies, follow-up period, outcome, and analysis plan) and main causal assumptions (i.e., conditional exchangeability, positivity, and consistency). Among the 2184 titles screened, 79 studies met the inclusion criteria. The methodologies used were G methods (61%) and reinforcement learning methods (39%). Studies considered both static (51%) and dynamic treatment regimes (49%). Only 30 (38%) of the studies reported all five target trial components, and only seven (9%) studies mentioned all three causal assumptions. To achieve actionable AI in the ICU, we advocate careful consideration of the causal question of interest, describing this research question as a target trial emulation, usage of appropriate causal inference methods, and acknowledgement (and examination of potential violations of) the causal assumptions.

Lung ultrasound to predict gas-exchange response to prone positioning in COVID-19 patients: A prospective study in pilot and confirmation cohorts.

PURPOSE: To examine whether lung ultrasound prior to prone positioning can predict the resulting gas-exchange response. MATERIALS AND METHODS: This is a prospective observational study on critically-ill COVID-19 patients with a pilot and confirmation cohort. Lung ultrasound examinations were performed before prone positioning and gas-exchange parameters were recorded before and after position change. RESULTS: A total of 79 patients, 36 in the pilot cohort and 43 in the confirmation cohort, were included. In the pilot cohort, a moderate correlation between pre-turn lung ultrasound score index (LUSI) and change in PaO2/FiO2 after prone positioning was found. These findings were corroborated and extended upon in the confirmation cohort. The confirmation cohort found that anterior LUSI had the strongest correlation with follow-up time-points 1, 6, 12, and 24 h after prone positioning, with strength of correlation gradually increasing up to 24 h. In a multivariate model anterior aeration loss (odds ratio 0.035; 95%CI 0.003-0.319 for anterior LUSI >50%) and higher pre-turn PaCO2 (odds ratio 0.479 95% CI 0.235-0.979) were negatively predictive of a PaO2/FiO2 increase ≥20 mmHg. CONCLUSIONS: Anterior LUSI, in addition to other clinical parameters, may be used to aid COVID-19 respiratory strategy and a clinician's decision to prone.

Dynamic prediction of mortality in COVID-19 patients in the intensive care unit: A retrospective multi-center cohort study.

Background: The COVID-19 pandemic continues to overwhelm intensive care units (ICUs) worldwide, and improved prediction of mortality among COVID-19 patients could assist decision making in the ICU setting. In this work, we report on the development and validation of a dynamic mortality model specifically for critically ill COVID-19 patients and discuss its potential utility in the ICU. Methods: We collected electronic medical record (EMR) data from 3222 ICU admissions with a COVID-19 infection from 25 different ICUs in the Netherlands. We extracted daily observations of each patient and fitted both a linear (logistic regression) and non-linear (random forest) model to predict mortality within 24 h from the moment of prediction. Isotonic regression was used to re-calibrate the predictions of the fitted models. We evaluated the models in a leave-one-ICU-out (LOIO) cross-validation procedure. Results: The logistic regression and random forest model yielded an area under the receiver operating characteristic curve of 0.87 [0.85; 0.88] and 0.86 [0.84; 0.88], respectively. The recalibrated model predictions showed a calibration intercept of -0.04 [-0.12; 0.04] and slope of 0.90 [0.85; 0.95] for logistic regression model and a calibration intercept of -0.19 [-0.27; -0.10] and slope of 0.89 [0.84; 0.94] for the random forest model. Discussion: We presented a model for dynamic mortality prediction, specifically for critically ill COVID-19 patients, which predicts near-term mortality rather than in-ICU mortality. The potential clinical utility of dynamic mortality models such as benchmarking, improving resource allocation and informing family members, as well as the development of models with more causal structure, should be topics for future research.

Multi-organ point-of-care ultrasound for detection of pulmonary embolism in critically ill COVID-19 patients - A diagnostic accuracy study.

PURPOSE: Critically ill COVID-19 patients have an increased risk of developing pulmonary embolism (PE). Diagnosis of PE by point-of-care ultrasound (POCUS) might reduce the need for computed tomography pulmonary angiography (CTPA), while decreasing time-to-diagnosis. MATERIALS & METHODS: This prospective, observational study included adult ICU patients with COVID-19. Multi-organ (lungs, deep vein, cardiac) POCUS was performed within 24 h of CTPA, looking for subpleural consolidations, deep venous thrombosis (DVT), and right ventricular strain (RVS). We reported the scan time, and calculated diagnostic accuracy measures for these signs separately and in combination. RESULTS: 70 consecutive patients were included. 23 patients (32.8%) had a PE. Median scan time was 14 min (IQR 11-17). Subpleural consolidations' diagnostic accuracy was: 42.9% (95%CI [34.1-52.0]). DVT's and RVS' diagnostic accuracy was: 75.6% (95%CI [67.1-82.9]) and 74.4% (95%CI [65.8-81.8]). Their sensitivity was: 24.0% (95%CI [9.4-45.1]), and 40.0% (95%CI [21.3-61.3]), while their specificity was: 88.8% (95%CI [80.8-94.3]), and: 83.0% (95%CI [74.2-89.8]), respectively. Multi-organ POCUS sensitivity was: 87.5% (95%CI [67.6-97.3]), and specificity was: 25% (95%CI [16.9-34.7]). CONCLUSIONS: Multi-organ rather than single-organ POCUS can be of aid in ruling out PE in critically ill COVID-19 and help select patients for CTPA. In addition, finding RVS can make PE more likely, while a DVT would preclude the need for a CTPA. REGISTRATION: www.trialregister.nl: NL8540.

Long-Term Changes in Free Nipple Graft Morphology and Patient-Reported Outcomes in Gender-Affirming Mastectomies.

BACKGROUND: After gender-affirming mastectomies with free nipple grafts, satisfaction with nipples tends to fall short behind chest outcomes. This might be related to changes in nipple areolar complex (NAC) dimensions over time. Therefore, the aim of this study is to establish the long-term changes in NAC morphology and compare these to cisgender male NAC outcomes. MATERIALS: An observational, cross-sectional study was performed. Data from two prospective cohorts were collected: (1) transgender men after a mastectomy with free nipple grafts and (2) cisgender men (reference sample). Demographics and 3-D images were collected for both groups. NAC measurements were performed on the 3-D images at 4 time points (7-, 30-, 90- and 365-days postoperative) in transgender men and once in cisgender men. Furthermore, the BODY-Q (nipple module) was administered postoperatively at 30-, 90- and 365-days in transgender men. RESULTS: In total, 67 transgender and 150 cisgender men were included. NAC width and height in trans men changed from 21.5 mm (±2.7) to 23.8 mm (±3.9, p<0.001) and 16.2 mm (±2.5) to 14.7 mm (±3.0, p=0.01) within a year, respectively. On average, the NACs increasingly rotated 21 degrees in the latero-caudal direction (p<0.001). The mean NAC width and height in cisgender men were 28.1 mm (±5) and 20.7 mm (±4), being significantly larger than in transgender men. Satisfaction for size, shape and flatness decreased significantly after postoperative day 30 (p=<0.05) in transgender men. CONCLUSIONS: Morphology of and satisfaction with the NACs in transgender men significantly changed over time. Understanding and incorporating these differences into pre-operative counseling and surgical planning might help increase patient satisfaction. LEVEL OF EVIDENCE IV: This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Within-patient randomized clinical trial comparing incisional negative-pressure wound therapy with suction drains in gender-affirming mastectomies.

BACKGROUND: Incisional negative-pressure wound therapy (iNPWT) is widely adopted by different disciplines for multiple indications. Questions about the most appropriate uses and value of iNPWT have been raised. METHODS: An open-label within-patient RCT was conducted in transgender men undergoing gender-affirming mastectomies. The objective was to determine the effect of iNPWT as a substitute for standard dressing and suction drains on wound healing complications. One chest side was randomized to receive the iNPWT intervention, and the other to standard dressing with suction drain. The primary endpoints were wound healing complications (haematoma, seroma, infection, and dehiscence) after three months. Additional outcomes were pain according to a numerical rating scale and patient satisfaction one week after surgery. RESULTS: Eighty-five patients were included, of whom 81 received both the iNPWT and standard treatment. Drain removal criteria were met within 24 h in 95 per cent of the patients. No significant decrease in wound healing complications was registered on the iNPWT side, but the seroma rate was significantly increased. In contrast, patients experienced both significantly less pain and increased comfort on the iNPWT side. No medical device-related adverse events were registered. CONCLUSION: Substituting short-term suction drains with iNPWT in gender-affirming mastectomies increased the seroma rates and did not decrease the amount of wound healing complications. Registration number: NTR7412 (Netherlands Trial Register).

Short-term cost-effectiveness of one-stage implant-based breast reconstruction with an acellular dermal matrix versus two-stage expander-implant reconstruction from a multicentre randomized clinical trial.

BACKGROUND: Implant-based breast reconstruction (IBBR) is the most commonly performed reconstructive procedure and its economic impact is significant. This study aimed to analyse whether a direct one-stage IBBR with use of an acellular dermal matrix (ADM) is more cost-effective than two-stage (expander-implant) breast reconstruction. METHODS: The BRIOS (Breast Reconstruction In One Stage) study was an open-label multicentre RCT in which women scheduled for skin-sparing mastectomy and immediate IBBR were randomized between one-stage IBBR with ADM or two-stage IBBR. Duration of surgery and hospital stay, and visits for the primary surgery, unplanned and cosmetic procedures were recorded. Costs were estimated at an institutional level. Health status was assessed by means of the EuroQol Five Dimensions 5L questionnaire. RESULTS: Fifty-nine patients (91 breasts) underwent one-stage IBBR with ADM and 62 patients (92 breasts) two-stage IBBR. The mean(s.d.) duration of surgery in the one-stage group was significantly longer than that for two-stage IBBR for unilateral (2·52(0·55) versus 2·02(0·35) h; P < 0·001) and bilateral (4·03(1·00) versus 3·25(0·58) h; P = 0·017) reconstructions. Costs were higher for one-stage compared with two-stage IBBR for both unilateral (€12 448 (95 per cent c.i. 10 722 to 14 387) versus €9871 (9373 to 10 445) respectively; P = 0·025) and bilateral (€16 939 (14 887 to 19 360) versus €13 383 (12 414 to 14 669); P = 0·002) reconstructions. This was partly related to the use of relatively expensive ADM. There was no difference in postoperative health status between the groups. CONCLUSION: One-stage IBBR with ADM was associated with higher costs, but similar health status, compared with conventional two-stage IBBR. Registration number: NTR5446 ( http://www.trialregister.nl).

Predictors of complications after direct-to-implant breast reconstruction with an acellular dermal matrix from a multicentre randomized clinical trial.

BACKGROUND: In the multicentre randomized trial BRIOS (Breast Reconstruction In One Stage), direct-to-implant (DTI) breast reconstruction with an acellular dermal matrix (ADM) was associated with a markedly higher postoperative complication rate compared with two-stage tissue expander/implant breast reconstruction. This study aimed to identify factors that contribute to the occurrence of complications after DTI ADM-assisted breast reconstruction. METHODS: Data were obtained from the BRIOS study, including all patients treated with DTI ADM-assisted breast reconstruction. Logistic regression analyses were performed to identify factors predictive of postoperative complications. RESULTS: Fifty-nine patients (91 breasts) were included, of whom 27 (35 breasts) developed a surgical complication. Reoperations were performed in 29 breasts (32 per cent), with prosthesis removal in 22 (24 per cent). In multivariable analyses, mastectomy weight was associated with complications (odds ratio (OR) 1·94, 95 per cent c.i. 1·33 to 2·83), reoperations (OR 1·70, 1·12 to 2·59) and removal of the implant (OR 1·55, 1·11 to 2·17). Younger patients (OR 1·07, 1·01 to 1·13) and those who received adjuvant chemotherapy (OR 4·83, 1·15 to 20·24) more frequently required reoperation. In univariable analyses, adjuvant radiotherapy showed a trend towards more complications (OR 7·23, 0·75 to 69·95) and removal of the implant (OR 5·12, 0·76 to 34·44), without reaching statistical significance. CONCLUSION: Breast size appeared to be the most significant predictor of complications in DTI ADM-assisted breast reconstruction. The technique should preferably be performed in patients with small to moderate sized breasts. Registration number: NTR5446 ( http://www.trialregister.nl).

Translating the ABC-02 trial into daily practice: outcome of palliative treatment in patients with unresectable biliary tract cancer treated with gemcitabine and cisplatin.

BACKGROUND: Biliary tract cancer (BTC) is an uncommon cancer with an unfavorable prognosis. Since 2010, the standard of care for patients with unresectable BTC is palliative treatment with gemcitabine plus cisplatin, based on the landmark phase III ABC-02 trial. This current study aims to evaluate the efficacy and safety of gemcitabine and cisplatin in patients with unresectable cholangiocarcinoma and gallbladder cancer in daily practice that meet the criteria for the ABC-02 trial in comparison to patients who did not. METHODS: Patients diagnosed with unresectable BTC between 2010 and 2015 with an indication for gemcitabine and cisplatin were included. We divided these patients into three groups: (I) patients who received chemotherapy and met the criteria of the ABC-02 trial, (II) patients who received chemotherapy and did not meet these criteria and (III) patients who had an indication for chemotherapy, but received best supportive care without chemotherapy. Primary outcome was overall survival (OS) and secondary outcome was progression-free survival (PFS). RESULTS: We collected data of 208 patients, of which 138 (66.3%) patients received first line chemotherapy with gemcitabine and cisplatin. Median OS of 69 patients in group I, 63 patients in group II and 65 patients in group III was 9.6 months (95%CI = 6.7-12.5), 9.5 months (95%CI = 7.7-11.3) and 7.6 months (95%CI = 5.0-10.2), respectively. Median PFS was 6.0 months (95%CI = 4.4-7.6) in group I and 5.1 months (95%CI = 3.7-6.5) in group II. Toxicity and number of dose reductions (p = .974) were comparable between the two chemotherapy groups. CONCLUSION: First-line gemcitabine and cisplatin is an effective and safe treatment for patients with unresectable BTC who do not meet the eligibility criteria for the ABC-02 trial. Median OS, PFS and treatment side effects were comparable between the patients who received chemotherapy (group I vs. group II).

Accuracy and reproducibility of fast fractional flow reserve computation from invasive coronary angiography.

Fractional flow reserve (FFR) guided percutaneous coronary intervention (PCI) is associated with favourable outcome compared with revascularization based on angiographic stenosis severity alone. The feasibility of the new image-based quantitative flow ratio (QFR) assessed from 3D quantitative coronary angiography (QCA) and thrombolysis in myocardial infarction (TIMI) frame count using three different flow models has been reported recently. The aim of the current study was to assess the accuracy, and in particular, the reproducibility of these three QFR techniques when compared with invasive FFR. QFR was derived (1) from adenosine induced hyperaemic coronary angiography images (adenosine-flow QFR [aQFR]), (2) from non-hyperemic images (contrast-flow QFR [cQFR]) and (3) using a fixed empiric hyperaemic flow [fixed-flow QFR (fQFR)]. The three QFR values were calculated in 17 patients who prospectively underwent invasive FFR measurement in 20 vessels. Two independent observers performed the QFR analyses. Mean difference, standard deviation and 95% limits of agreement (LOA) between invasive FFR and aQFR, cQFR and fQFR for observer 1 were: 0.01 ± 0.04 (95% LOA: -0.07; 0.10), 0.01 ± 0.05 (95% LOA: -0.08; 0.10), 0.01 ± 0.04 (95% LOA: -0.06; 0.08) and for observer 2: 0.00 ± 0.03 (95% LOA: -0.06; 0.07), -0.01 ± 0.03 (95% LOA: -0.07; 0.05), 0.00 ± 0.03 (95% LOA: -0.06; 0.05). Values between the 2 observers were (to assess reproducibility) for aQFR: 0.01 ± 0.04 (95% LOA: -0.07; 0.09), for cQFR: 0.02 ± 0.04 (95% LOA: -0.06; 0.09) and for fQFR: 0.01 ± 0.05 (95% LOA: -0.07; 0.10). In a small number of patients we showed good accuracy of three QFR techniques (aQFR, cQFR and fQFR) to predict invasive FFR. Furthermore, good inter-observer agreement of the QFR values was observed between two independent observers.

Autologous fat grafting in onco-plastic breast reconstruction: A systematic review on oncological and radiological safety, complications, volume retention and patient/surgeon satisfaction.

OBJECTIVE: This study presents an up-to-date overview of the literature on autologous fat grafting (AFG) in onco-plastic breast reconstruction, with respect to complications, oncological and radiological safety, volume retention and patient/surgeon satisfaction. BACKGROUND: Although AFG is increasingly being applied in onco-plastic breast reconstruction, a comprehensive overview of the available evidence for this procedure is still lacking. METHODS: A systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement was conducted. Case series, cohort studies and randomized controlled trials (RCTs) reporting on relevant outcomes of breast reconstruction with supplemental AFG were included. RESULTS: In total, 43 studies were included reporting on 6260 patients with a follow-up period ranging from 12 to 136 months. The average locoregional and distant oncological recurrence rates after breast reconstruction with AFG were 2.5% (95% confidence interval (CI) 1.7-3.7) and 2.0% (95% CI 1.1-3.5), respectively. Fewer cysts and calcifications were seen on radiological images for this procedure than for other types of breast surgery. However, more biopsies were performed based on radiological findings (3.7% vs. 1.6%), and more cases of fat necrosis (9.0% vs 4.7%) were seen after treatment with AFG. The total complication rate of 8.4% (95% CI 7.6-9.1) is lower than those reported following other reconstructive breast procedures. The mean volume retention was 76.8% (range 44.7-82.6%) with a satisfaction rate of 93.4% for patients and 90.1% for surgeons. CONCLUSIONS: AFG in breast reconstruction is a promising technique. Safety is not compromised as cancer recurrence and complications are not observed. Whether AFG interferes with radiological follow-up remains to be elucidated. Randomized trials with sound methodology are needed to confirm these conclusions.