Literature Search
In accordance with the PRISMA guidelines, our two investigators (JJ-C, G-K) systematically and independently conducted a review of the relevant published data. A computerized search of the Pubmed, Embase, MEDLINE and Cochrane electronic databases was performed using the keywords “furosemide,” “furosemide stress test,” “acute kidney injury,” “acute kidney failure,” and “acute renal insufficiency,” as well as the medical subject heading (MeSH) terms "Furosemide" [Mesh] AND "Acute Kidney Injury" [Mesh], in order to identify all the relevant English-language studies up to December 2019. Review articles or meta-analyses were not included for analysis, but their citations and references were searched for additional relevant studies.
Study Selection
After the initial screening, the two investigators Jia Jin Chen (JJ-C) and George Kuo (G-K) independently determined the eligibility of the identified studies based on evaluations of their titles, abstracts, and, subsequently, full texts. Any difference in opinion regarding eligibility was resolved by consensus through discussion. The full text of any article that was deemed potentially relevant was retrieved online. A study was included if it met the criteria of being a published English-language study, adult humans as its population, and reported the protocol and cut-off point of the FST. We enrolled studies with primary or secondary outcomes reporting the diagnostic value of the FST for AKI progression, RRT, or mortality. Studies were excluded if they met one or more of the following criteria: (1) focused on a population with transplanted kidneys, (2) used duplicate cohorts, (3) contained insufficient information for analysis, (4) were based on a child population, or (5) included no reported outcome of interest. Detailed results regarding excluded studies and the reasons for their exclusion are available in Supplemental Table 1. We have registered our work in PROSPERO with the study ID number 160934. However, till we finished our work, the registration was still under assessed by the editorial team of PROSPERO.
Data Extraction
The two investigators independently extracted relevant information from each study. The extracted data elements related to the study-level characteristics included the first author, year of publication, study location, study design, definition of AKI, total sample size, protocol of the FST (that is, furosemide dose, time interval, cut-off point urine output), enrolled patients’ AKI stages, reported outcomes of interest, whether or not the enrolled population had high plasma neutrophil gelatinase-associated lipocalin (NGAL) levels, and whether or not patients with chronic kidney disease were excluded (Table 1). As for diagnostic test performance, the extracted data included the cut-off point urine output based on the Youden index or pre-defined criteria, sensitivity, specificity, value of area under the receiver operating characteristics (AUROC), and the event number of AKI or RRT or mortality (Table 1 and Table 2).
Table 1
The characteristics of the nine included studies
First author/ year | Location | Design | AKI criteria | Population | Sample size | Furosemide dose | Urine output cutoff point | Outcome of interest | Enrolled patients AKI stage | High plasma NGAL | Exclusion of chronic kidney disease |
Chawla, 2013 | USA | PC + RC | AKIN | Mixed | 77 | 1 mg/kg (furosemide naïve) or 1.5 mg/kg (furosemide non-naïve) | 200 ml/ 2hr | AKIN stage 3 | AKIN stage 1–2 | Yes | Yes (eGFR < 30) |
Elsaegh, 2018 | Egypt | PC | KDIGO | Sepsis | 60 | KDIGO stage progression (included RRT) | Normal renal function & any stage of AKI | No | Yes |
Lumlertgul, 2018 | Thailand | PC | KDIGO | Mixed | 162 | RRT | Any stage of AKI | Yes | Yes (baseline Cr > 2) |
Matsuura, 2018 | Japan | RC | KDIGO | Mixed | 51 | NR | 3.9 ml/2hr for per mg furosemide | KDIGO stage 3 & RRT | KDIGO stage 1–2 or high NGAL with normal renal function | Yes | No |
Saber, 2018 | Egypt | PC | AKIN | NR | 40 | 1 mg/kg (furosemide naïve) or 1.5 mg/kg (furosemide non-naïve) | 325 ml/ 6hr | AKIN stage 3 (included RRT) | AKIN stage 1–2 | No | Yes (eGFR < 30) |
Rewa, 2019 | USA and Canada | PC | AKIN | Mixed | 92 | 200 ml/ 2hr | AKIN stage 3 | AKIN stage 1–2 | No | Yes (eGFR < 30) |
Sakhuja, 2019 | USA | RC | AKIN | NR | 687 | ≥ 1 mg/kg | 600 ml/ 6hr | RRT | AKIN stage 3 | No | No |
Vairakkani, 2019 | India | NR | KDIGO | NR | 80 | 1 mg/kg (furosemide naïve) or 1.5 mg/kg (furosemide non-naïve) | 325 ml/ 2hr | KDIGO stage 3 | KDIGO stage 1–2 | No | Yes (eGFR < 30) |
Venugopal, 2019 | India | PC | AKIN | NR | 62 | 200 ml/ 2hr | AKIN-3 & RRT | AKIN stage 1–2 | No | No |
Abbreviation: AKI (acute kidney injury), AKIN (Acute Kidney Injury Network), Cr (Creatinine), eGFR (estimated Glomerular filtration rate), KDIGO (Kidney Disease Global outcomes), NR (not report), PC (prospective cohort), RC (Retrospective cohort), RRT (Renal replacement therapy) |
Table 2
Diagnostic test performance of furosemide stress test for AKI progression ,renal replacement therapy and mortality
Study | Sensitivity | Specificity | AUROC | sample size | Event (AKI progression) | TP | FP | FN | TN | Follow up period |
Chawla, 2013 | 87.1 | 84.1 | 0.87 | 77 | 25 | 22 | 8 | 3 | 44 | 14 days |
Elsaegh, 2018 | 89.3 | 93.4 | NR | 60 | 28 | 25 | 2 | 3 | 30 | NR |
Matsuura, 2018 | 76.5 | 94.1 | 0.84 | 51 | 17 | 13 | 2 | 4 | 32 | 7 days |
Saber, 2018 | 86.7 | 68 | NR | 40 | 15 | 13 | 8 | 2 | 17 | NR |
Rewa, 2019 | 73.9 | 90 | 0.87 | 92 | 23 | 17 | 7 | 6 | 62 | 30 days |
Vairakkani, 2019 | 82 | 80.8 | NR | 80 | 28 | 23 | 10 | 5 | 42 | 14 days |
Venugopal, 2019 | 85.7 | 87.5 | NR | 62 | 14 | 12 | 6 | 2 | 42 | NR |
Study | Sensitivity | Specificity | AUROC | sample size | Event (RRT) | TP | FP | FN | TN | Follow up period |
Lumlertgul, 2018 | 94.4 | 70.4 | NR | 94 | 108 | 102 | 16 | 6 | 38 | NR |
Matsuura, 2018 | 75 | 79 | NR | 51 | 8 | 6 | 9 | 2 | 34 | 7 days |
Sakhuja, 2019 | 80.9 | 50.5 | NR | 687 | 162 | 131 | 260 | 31 | 265 | 1 days |
Venugopal, 2019 | 83.3 | 84 | NR | 62 | 12 | 10 | 8 | 2 | 42 | NR |
Study | Sensitivity | Specificity | AUROC | sample size | Event (mortality) | TP | FP | FN | TN | Follow up period |
Venugopal, 2019 | 66.7 | 77/3 | NR | 62 | 9 | 6 | 12 | 3 | 41 | NR |
Abbreviation: AUROC (Area Under the Receiver Operating Characteristics), AKI (Acute kidney injury), FN (False negative), FP (False positive), NR (not report), RRT (Renal replacement therapy), TN (True negative), TP (True positive) |
Outcome Measures
The diagnostic criteria for AKI were different in the nine enrolled studies. Four of the studies (Elsaegh, 2018; Lumlertgul, 2018; Matsuura, 2018; Vairakkani, 2019)[14–17] used the Kidney Disease: Improving Global Outcomes (KDIGO) criteria [18]. Other studies used the Acute Kidney Injury Network (AKIN) criteria [19]. The reference test used in each study was based on the different AKI criteria used in each trial or on whether the patients received RRT. Four studies (Chawla, 2013; Rewa, 2019; Saber, 2018; Venugopal, 2019)[8, 20–22] used the AKIN stage 3 AKI criteria. Three studies (Elsaegh, 2018; Matsuura, 2018; Vairakkani, 2019)[14, 16–17] used the KDIGO stage 3 AKI criteria.
Risk of Bias Assessment
The risk of bias for each of the included studies was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool and Review Manager version 5.3 to identify the quality of the included studies [23]. The QUADAS-2 tool is based on four domains (patient selection, index test, reference standard, and flow and timing), which are used to judge the risk of bias. Each study was reviewed independently by JJ-C and G-K, with each investigator assigning a rating of high, low, or unclear risk for all four domains. The judgment principle of “applicability” was the same as the bias section, but there were no signaling questions. Disagreements between the reviewers were resolved by discussion. If the answer to all the signaling questions for a given domain was “yes,” then the domain was considered to entail a low risk of bias. If the answer to any of the signaling questions for a domain was “no,” then the domain was considered to entail a high risk of bias. The quality of evidence for the diagnostic performance of the FST in this meta-analysis was assessed based on the guidelines of the GRADE Working Group methodology [41]. We summarized the results in a table, which was constructed using the online GRADE Profiler (see Supplementary Table 2).
Statistical Analysis
We extracted the event number, total sample size, and true positive (TP), true negative (TN), false positive (FP), and false negative (FN) rates for each study or calculated these values according to the reported sensitivity and specificity. Based on these data, the positive likelihood ratio (+ LR), negative likelihood ratio (-LR), and diagnostic odds ratio (DOR) could be obtained for each study. The summary measures were calculated using a bivariate model for the obtained pooled sensitivity and specificity. We used a random-effect model with maximum likelihood estimation to calculate the pooled DOR and LR. The above two tests were conducted by the ‘metabin’ function in the ‘meta’ package [24]. To assess the diagnostic performance of the FST regarding AKI progression for FST non-responders, a summary receiver operating characteristics (SROC) curve was constructed by the ‘restima’ function with restricted maximum likelihood estimation in the ‘mada’ package [42]. The threshold effect was examined by using the Spearman correlation coefficient between the logit of sensitivity and logit of “1 – specificity,” and P < 0.05 indicated the existence of a threshold effect. If there is no significant threshold effect, subgroup analysis or meta-regression analysis is warranted to clarify the sources of heterogeneity [25]. Heterogeneity from covariates other than the threshold effect among studies was evaluated using the I2 index, with I2 < 25%, 25% − 50%, and > 50% indicating mild, moderate, and high heterogeneity, respectively. The LRs indicate whether the accuracy of a particular test would be more accurate for patients with a disease than for subjects without the disease. Several relevant variables were identified, and these variables are summarized in Table 1 and Table 2 (with the specific variables including the AKI criteria used, whether or not the enrolled patients had high plasma NGAL, whether or not the enrolled patients had a clinical diagnosis of AKI, the pre-specific urine output cut-off point, the study design, and whether or not patients with chronic kidney disease were excluded). To explore possible sources of heterogeneity, these variables were applied as moderators in meta-regression weighted by the inverse of the study variance. We performed the meta-regression by using Meta-DiSc (version 1.4) [26]. A sensitivity analysis was performed after excluding studies used the composite outcomes of AKI stage progression and RRT. All analyses were conducted using R version 3.6.2 (2019-12-12) [43]. A two-sided P value of < 0.05 was considered statistically significant.