Abstract
The blood urea nitrogen (BUN) and blood urea nitrogen-to-creatinine ratio (BCR) are elevated in patients with upper gastrointestinal bleeding (UGIB). If bleeding persists after an initial endoscopic hemostasis, BUN levels will remain elevated or continue to increase. When identifying the occurrence of rebleeding is difficult, clinicians should perform second-look endoscopy, which is costly and invasive. This study aimed to evaluate whether changes in the BCR or BUN levels aid in determining rebleeding in patients with UGIB. Two hundred eleven patients underwent an initial and follow-up endoscopy within 72 hours for non-variceal UGIB at Chungbuk National University Hospital (Cheongju, Korea). We evaluated the changes in the BCR and BUN levels, based on the presence of rebleeding within 72 hours of the initial endoscopy. Twenty-five (11.8%) patients had rebleeding at the follow-up endoscopy within 72 hours after the initial endoscopy. The overall initial BCR was 38.3 ± 16.0 (median: 35.8). The BCR difference at follow-up endoscopy was 14.1 ± 14.5 in the no-rebleeding group and 4.5 ± 13.2 in the rebleeding group (P = .004). The optimal cutoff value of the BCR difference to predict rebleeding was 8.8. The changes in the BUN level at the follow-up endoscopy were 14.4 ± 12.3 mg/dL in the no-rebleeding group and 5.4 ± 9.5 mg/dL in the rebleeding group (P = .001). The optimal cutoff value of the BUN difference that predicted rebleeding was 7.5 mg/dL. Multivariate analysis revealed that older patients tended to have more rebleeding (P = .034). BCR and BUN level changes may aid in determining whether rebleeding has occurred, thereby preventing unnecessary second-look endoscopy in patients with UGIB.
Keywords: blood urea nitrogen, rebleeding, upper gastrointestinal bleeding
1. Introduction
Upper gastrointestinal bleeding (UGIB) is a hemorrhage in the gastrointestinal tract proximal to the Treitz ligament.[1,2] UGIB remains a potentially life-threatening disease, although the incidence and mortality of UGIB have decreased because of the development of endoscopic and assisted therapies.[1,3] Even when endoscopic hemostasis is successful, approximately 10% to 15% of patients develop rebleeding [4].
In UGIB, the initial management and evaluation of rebleeding are important for patient prognosis.[5] Various predictors such as history, blood pressure, heart rate, laboratory tests, and nasogastric lavage have been used to determine the occurrence of rebleeding. However, no clear indicators of rebleeding exist, unless bleeding is severe. The most obvious method is to perform second-look endoscopy; however, this process is costly and invasive. Second-look endoscopy is not routinely recommended because it can be an additional burden to patients and physicians.[6] Therefore, identifying surrogate markers that can predict rebleeding is necessary.
Blood urea nitrogen (BUN) levels increase during gastrointestinal bleeding, particularly, in patients with UGIB. However, its levels can increase in a variety of other conditions, including acute kidney injury, high-protein diet, dehydration, and aging.[7–10] By contrast, plasma creatinine (Cr) does not change with age because of the reduction in Cr production and urinary Cr output.[10] Therefore, the blood urea nitrogen-to-creatinine ratio (BCR) is usually used to increase the diagnostic value of gastrointestinal bleeding, and BCR could be used to differentiate between acute or chronic renal disease and gastrointestinal bleeding. In addition, previous studies[3,11,12] have demonstrated that the BCR can be used as an indicator to distinguish whether the origin of gastrointestinal bleeding is the upper or lower gastrointestinal tract and that patients with high BCR have poor outcomes. To date, no study has evaluated the role of the BCR as a predictor of rebleeding in patients with UGIB. Therefore, we evaluated whether the change in the BCR within 72 hours after the initial endoscopy would aid in determining whether rebleeding has occurred in patients with non-variceal UGIB.
2. Methods
2.1. Study population
In total, the medical records of 1081 patients who underwent endoscopic hemostasis for non-variceal UGIB at Chungbuk National University Hospital (Cheongju, Republic of Korea) between January 2010 and June 2019 were reviewed. We included patients who underwent follow-up endoscopy within 72 hours and had laboratory findings such as the levels of hemoglobin, BUN, and Cr at their initial and follow-up visits. All patients were >20 years of age. We excluded patients with chronic kidney disease, malignancy of the gastrointestinal tract, or a medical history of gastrointestinal surgery, including bowel resection. All patients received high-dose intravenous proton-pump inhibitors (80 mg bolus, followed by 8 mg per hour) on admission and for 72 hours. Endoscopic hemostasis was performed using various modalities, including hemoclipping, thermal coagulation, injection therapy, band ligation, or combinations thereof, at the discretion of the endoscopist.
2.2. Data collection
We reviewed the patients’ medical records and collected their data, which included demographics, comorbidities, and medications such as antithrombotics. We also collected data on vital signs and laboratory and endoscopic findings at the initial and follow-up visits. Rebleeding was defined as overt bleeding, such as spurting or an oozing hemorrhage at follow-up endoscopy within 72 hours of the initial endoscopic hemostasis. The initial laboratory variables were measured before the index endoscopy. Follow-up BUN levels, which were used to evaluate longitudinal changes in BUN and BCR, were measured after the index endoscopic hemostasis and before the second-look endoscopy. At our center, second-look endoscopy was routinely performed within 72 hours after initial hemostasis, regardless of clinical signs or laboratory changes, as part of standard post-hemostasis care.
2.3. Outcome assessment
The primary goal of this study was to evaluate whether a change in the BCR within 72 hours after the initial endoscopy would aid in determining the presence of rebleeding in patients with non-variceal UGIB. The secondary objective was to identify risk factors and other predictors of rebleeding.
2.4. Statistical analysis
Descriptive statistics were used to analyze the clinical characteristics of the patients. Categorical data were presented as numbers (percentages), and continuous data were presented as mean (standard deviation) or median (interquartile range [IQR]), as appropriate. Clinical and demographic variables were analyzed by using the chi-square or Fisher’s exact test for categorical variables and the t-test or Mann–Whitney U test for continuous variables. We calculated and interpreted receiver operating characteristic curves to select the optimal cutoff values for differences in the BCR and BUN levels. Statistical significance was defined as P < .05, and all statistical analyses and calculations were conducted using R (R Foundation for Statistical Computing, Vienna, Austria) and the statistical package for the social sciences for Windows, version 24.0 (SPSS Inc., Chicago).
2.5. Ethics statement
The study protocol was approved by the Institutional Review Board of the Chungbuk National University Hospital (application no. 2021-01-022). Informed consent was waived because of the retrospective nature of the study, and the analysis used anonymous clinical data. This study was conducted by the Declaration of Helsinki.
3. Results
3.1. Patients’ characteristics
In total, 211 patients who underwent endoscopic hemostasis and follow-up endoscopy within 72 hours for non-variceal UGIB were included in this study. Twenty-five (11.8%) patients had rebleeding on follow-up endoscopy. Gastric ulcers were the most common cause of non-variceal UGIB in both groups (51.1% in the no-rebleeding group, 48.0% in the rebleeding group). The rebleeding group was older than the no-rebleeding group (69.2 ± 13.9 years vs 58.5 ± 16.6 years, P = .034), and had lower serum albumin levels (3.2 ± 0.7 g/dL vs 3.5 ± 0.6 g/dL, P = .039). The median time interval from the initial endoscopy to second-look endoscopy was 22.0 hours (IQR: 18.0 to 26.0) in the no-rebleeding group and 23.0 hours (IQR: 18.0–41.0) in the rebleeding group, with no significant difference between the groups (P = .371, Table 1). The number of red blood cell transfusion units was significantly higher in the rebleeding group. The median number of transfused red blood cell units was 1.0 (IQR: 0.0–1.75) in the no-rebleeding group and 1.0 (IQR: 1.0–4.0) in the rebleeding group (P = .024, Table 1). Other clinical characteristics, including comorbidities, use of antithrombotic agents, blood pressure, pulse rate, and initial hemoglobin level, were not significantly different between the 2 groups. Multivariate analysis revealed that age was the only independent variable significantly associated with rebleeding (P = .034). A summary of these clinical and laboratory characteristics according to rebleeding status is presented in Table 1.
Table 1.
Clinical characteristics of patients with non-variceal UGIB without rebleeding and with rebleeding within 72 hours.
| Variable | No-rebleeding (n = 186) | Rebleeding (n = 25) | P-value |
|---|---|---|---|
| Age, yr, (mean ± SD, yr) | 58.5 ± 16.6 | 69.2 ± 13.9 | .001 |
| Sex (male:female) | 154 (82.8%):32 (17.2%) | 17 (68.0%):8 (32.0%) | .100 |
| Comorbidity | 73 (39.2%) | 13 (52.0%) | .279 |
| Hypertension | 56 (30.1%) | 11 (44.0%) | .175 |
| Diabetes | 33 (17.7%) | 7 (28.0%) | .274 |
| Antithrombotics* | 53 (28.5%) | 7 (28.0%) | 1.000 |
| Bleeding cause | .958 | ||
| Gastric ulcer | 95 (51.1%) | 12 (48.0%) | |
| Duodenal ulcer | 62 (33.3%) | 9 (36.0%) | |
| Other diseases† | 29 (15.6%) | 4 (16.0%) | |
| RBC transfusion | 120 (65.2%) | 19 (76.0%) | .369 |
| Number of RBC units transfused (median) | 1.0 (0.0–1.75) | 1.0 (1.0–4.0) | .024 |
| Time to second-look endoscopy (median, h) | 22.0 (18.0–26.0) | 23.0 (18.0–41.0) | .371 |
| SBP (mm Hg) | 112.7 ± 22.9 | 116.8 ± 33.4 | .913 |
| Pulse rate (/min) | 95.2 ± 17.8 | 90.4 ± 17.1 | .208 |
| Hemoglobin level (g/dL) | 9.7 ± 2.8 | 9.2 ± 2.9 | .501 |
| Platelet level (/µL) | 224,419.4 ± 76,238.2 | 194,920.0 ± 94,932.0 | .161 |
| Prothrombin time (INR) | 1.24 ± 0.68 | 1.24 ± 0.32 | .083 |
| Albumin level (g/dL) | 3.5 ± 0.6 | 3.2 ± 0.7 | .039 |
Data are presented as the mean ± the standard deviation or as the number (%).
INR = international normalized ratio, RBC = red blood cell, SBP = systolic blood pressure, SD = standard deviation, UGIB = upper gastrointestinal bleeding.
Antithrombotics include antiplatelet and anticoagulant agents.
Other diseases include Mallory–Weiss syndrome, gastric or duodenal erosion, and angiodysplasia.
3.2. Endoscopic hemostasis methods
A total of 7 categories of endoscopic hemostasis were identified among the included patients. The most commonly used method was hemoclipping (n = 68), followed by a combination of hemoclip and fibrin glue injection (n = 45), hemoclip and epinephrine injection (n = 33), fibrin glue injection alone (n = 28), band ligation (n = 24), epinephrine injection alone (n = 11), and thermal coagulation (n = 2). Table 2 summarizes the distribution of these treatment modalities according to the no-rebleeding and rebleeding groups.
Table 2.
Endoscopic hemostasis techniques applied during initial UGIB management.
| Endoscopic hemostasis method | No-rebleeding (n = 186) | Rebleeding (n = 25) |
|---|---|---|
| Hemoclip | 58 | 10 |
| Fibrin glue injection | 24 | 4 |
| Epinephrine injection | 11 | 0 |
| Band ligation | 22 | 2 |
| Thermal coagulation | 0 | 2 |
| Hemoclip with epinephrine injection | 33 | 0 |
| Hemoclip with fibrin glue injection | 38 | 7 |
UGIB = upper gastrointestinal bleeding.
3.3. Study outcomes
The overall initial BCR was 38.3 ± 16.0 (median: 35.8). Changes in Cr levels did not differ between the 2 groups (P = .480), whereas changes in the BUN level and BCR were significantly different (P = .001 and P = .004, respectively) (Fig. 1A–C). The mean change in the BUN levels at the follow-up endoscopy within 72 hours was 14.4 ± 12.3 mg/dL in the no-rebleeding group and 5.4 ± 9.5 mg/dL in the rebleeding group (P = .001). In addition, the change in the BCR at the follow-up endoscopy within 72 hours was 14.1 ± 14.5 in the no-rebleeding group and 4.5 ± 13.2 in the rebleeding group (P = .004) (Table 3). Based on the receiver operating characteristic curves, the optimal cutoff values for determining the presence of rebleeding were 8.8 for the BCR difference with a sensitivity of 61.3% and specificity 64.0%, and 7.5 mg/dL for the BUN difference with a sensitivity of 66.7% and specificity 64.0% (Fig. 2A–B).
Figure 1.
(A) changes in the Cr levels, based on the presence of rebleeding. (B) Change in the BUN levels, based on the presence of rebleeding. (C) Changes in the BCR, based on the presence of rebleeding. BCR = blood urea nitrogen-to-creatinine ratio, BUN = blood urea nitrogen, Cr = creatinine.
Table 3.
Comparison of changes in the Cr and BUN levels and the BCR between patients with non-variceal UGIB without rebleeding and with rebleeding.
| Variable | No-rebleeding (n = 186) | Rebleeding (n = 25) | P-value |
|---|---|---|---|
| Initial Cr level (mg/dL) | 0.96 ± 0.26 | 0.94 ± 0.35 | .468 |
| FU Cr level (mg/dL) | 0.92 ± 0.21 | 0.87 ± 0.27 | .300 |
| Cr difference | 0.05 ± 0.17 | 0.04 ± 0.23 | .480 |
| Initial BUN level (mg/dL) | 36.0 ± 15.6 | 34.0 ± 17.8 | .295 |
| FU BUN level (mg/dL) | 21.6 ± 10.8 | 29.8 ± 18.2 | .017 |
| BUN difference | 14.4 ± 12.3 | 5.4 ± 9.5 | .001 |
| Initial BCR | 38.3 ± 15.9 | 37.9 ± 16.6 | .853 |
| FU BCR ratio | 24.2 ± 11.8 | 35.1 ± 16.9 | .001 |
| BCR difference | 14.1 ± 14.5 | 4.5 ± 13.2 | .004 |
Data are presented as the mean ± standard deviation.
BCR = blood urea nitrogen-to-creatinine ratio, BUN = blood urea nitrogen, Cr = creatinine, FU = follow-up, UGIB = upper gastrointestinal bleeding.
Figure 2.
(A) ROC curve of the BCR difference based on the presence of rebleeding. (B) ROC curve of the BUN difference based on the presence of rebleeding. AUC = area under the curve, BCR = blood urea nitrogen-to-creatinine ratio, BUN = blood urea nitrogen, CI = confidence interval, ROC = receiver operating characteristic.
4. Discussion
This retrospective study aimed to evaluate whether changes in BCR or BUN levels would aid in determining the presence of rebleeding in patients with UGIB. Among 211 patients with non-variceal UGIB, we identified that changes in the BCR and BUN may be predictors of successful hemostasis in patients with UGIB. The decrease in the BUN level and BCR was significantly smaller in patients with rebleeding than in patients without rebleeding (BUN: P = .004; BCR: P = .001). This finding may be because BUN levels increase as blood is absorbed in the small intestine if bleeding persists, whereas no blood is absorbed if bleeding stops. The BUN level also increases in cases of dehydration, a high-protein diet, and acute kidney injury. Therefore, we initially considered the BCR as a predictive marker of rebleeding to exclude factors associated with BUN emission problems in the kidney. However, in our study, we found that changes in the BCR were caused by a decrease in the BUN level, rather than changes in the Cr level (Fig. 1 and Table 3). Therefore, changes in BUN levels may be a good surrogate marker for rebleeding.
Previous studies[13–15] have demonstrated that the BCR is an indicator of the origin of gastrointestinal bleeding (i.e., whether the bleeding focus is in the upper or lower gastrointestinal tract). Its diagnostic value is not great, although it has a sensitivity of 90%, specificity of 27%, and an area under the curve of 0.73.[13] The findings of these studies suggest that UGIB is highly likely if the BCR is >35 to 36. In our study, the initial BCR was 38.3 ± 16.0 (median: 35.8). Because all patients included in our study had UGIB, this value is consistent with the results of previous studies.[13,15] However, previous studies[13–15] did not assess the BCR as a predictor of rebleeding, despite identifying the BCR as a useful tool for determining the origin of bleeding. Based on our findings, the BCR may also be an indicator of rebleeding after endoscopic hemostasis in patients with UGIB.
We expected rebleeding to be more common in patients who had comorbidities or were taking antithrombotics. However, no significant difference existed in the occurrence of rebleeding in such patients. In the multivariate analysis, age was the only significant clinical factor between rebleeding and non-rebleeding patients (P = .034).
Our study had several limitations. First, it was a retrospective study and not a randomized prospective study. Therefore, some data were missing, and risk factors or other predictors of rebleeding could not be thoroughly evaluated. Second, the number of patients with rebleeding was small because all patients included in our study underwent an initial endoscopic hemostasis. Third, we could not identify the reason for the greater occurrence of rebleeding among the older patients in our study. Older patients are more likely to have comorbidities and to take drugs that can cause UGIB, such as nonsteroidal anti-inflammatory drugs, selective serotonin reuptake inhibitors, and calcium channel blockers.[16–20] In addition, this may be because of the lack of the ability to repair tissues with aging.[21]
To the best of our knowledge, no study has evaluated changes in the BCR or BUN levels as predictors of rebleeding in patients with UGIB. 1 study revealed an association between changes in BUN levels and prognosis[11] and suggested that elevated BUN levels are associated with a poor prognosis. However, the authors of the aforementioned study did not find an association between increased BUN and rebleeding. In addition, the presence of rebleeding in all patients in the present study was well-documented because all patients underwent follow-up endoscopy and had laboratory findings at their initial and follow-up examinations.
In conclusion, changes in the BCR and BUN levels may aid in determining the presence of rebleeding after initial endoscopic hemostasis, thereby preventing unnecessary second-look endoscopy in patients with UGIB. In particular, changes in the BCR and BUN levels are useful if rebleeding is ambiguous, endoscopy is difficult to implement, or the risk of endoscopic adverse effects is a concern, such as perforation, bleeding, adverse reactions to sedatives, and old age. Further randomized controlled trials are required to confirm the results of this study.
Author contributions
Data curation: Jun Su Lee, Ki Bae Kim.
Formal analysis: Jun Su Lee, Soon Man Yoon.
Conceptualization: Soon Man Yoon.
Writing – original draft: Jun Su Lee.
Writing – review & editing: Ki Bae Kim, Soon Man Yoon.
Abbreviations:
- BCR
- blood urea nitrogen-to-creatinine ratio
- BUN
- blood urea nitrogen
- Cr
- creatinine
- IQR
- interquartile range
- UGIB
- upper gastrointestinal bleeding
The authors have no funding and conflicts of interest to disclose.
The datasets generated during and/or analyzed during the current study are not publicly available, but are available from the corresponding author on reasonable request.
How to cite this article: Lee JS, Kim KB, Yoon SM. Use of the blood urea nitrogen level to predict the rebleeding in patients with non-variceal upper gastrointestinal bleeding. Medicine 2026;105:7(e47558).
Contributor Information
Jun Su Lee, Email: zzzgggh@naver.com.
Ki Bae Kim, Email: kibae@hanmail.net.
References
- [1].Khamaysi I, Gralnek IM. Acute upper gastrointestinal bleeding (UGIB) - initial evaluation and management. Best Pract Res Clin Gastroenterol. 2013;27:633–8. [DOI] [PubMed] [Google Scholar]
- [2].Lanas A, Dumonceau JM, Hunt RH, et al. Non-variceal upper gastrointestinal bleeding. Nat Rev Dis Primers. 2018;4:18020. [DOI] [PubMed] [Google Scholar]
- [3].Hajiagha Mohammadi AA, Reza Azizi M. Prognostic factors in patients with active non-variceal upper gastrointestinal bleeding. Arab J Gastroenterol. 2019;20:23–7. [DOI] [PubMed] [Google Scholar]
- [4].Park SJ, Park H, Lee YC, et al. Effect of scheduled second-look endoscopy on peptic ulcer bleeding: a prospective randomized multicenter trial. Gastrointest Endosc. 2018;87:457–65. [DOI] [PubMed] [Google Scholar]
- [5].Rodriguez de Santiago E, Burgos-Santamaria D, Perez-Carazo L, et al. ; TC-325 Collaboration Project, Endoscopy Group of the Spanish Association of Gastroenterology. Hemostatic spray powder TC-325 for GI bleeding in a nationwide study: survival and predictors of failure via competing risks analysis. Gastrointest Endosc. 2019;90:581–90.e6. [DOI] [PubMed] [Google Scholar]
- [6].Hwang JH, Fisher DA, Ben-Menachem T, et al. ; Standards of Practice Committee of the American Society for Gastrointestinal Endoscopy. The role of endoscopy in the management of acute non-variceal upper GI bleeding. Gastrointest Endosc. 2012;75:1132–8. [DOI] [PubMed] [Google Scholar]
- [7].Haugen HN. The significance of azotemia in gastro-intestinal hemorrhages. Scand J Clin Lab Invest. 1955;7:75–9. [DOI] [PubMed] [Google Scholar]
- [8].Stellato T, Rhodes RS, McDougal WS. Azotemia in upper gastrointestinal hemorrhage. A review. Am J Gastroenterol. 1980;73:486–9. [PubMed] [Google Scholar]
- [9].Wang Z, Li L, Glicksberg BS, Israel A, Dudley JT, Ma’ayan A. Predicting age by mining electronic medical records with deep learning characterizes differences between chronological and physiological age. J Biomed Inform. 2017;76:59–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Musch W, Verfaillie L, Decaux G. Age-related increase in plasma urea level and decrease in fractional urea excretion: clinical application in the syndrome of inappropriate secretion of antidiuretic hormone. Clin J Am Soc Nephrol. 2006;1:909–14. [DOI] [PubMed] [Google Scholar]
- [11].Kumar NL, Claggett BL, Cohen AJ, Nayor J, Saltzman JR. Association between an increase in blood urea nitrogen at 24 hours and worse outcomes in acute nonvariceal upper GI bleeding. Gastrointest Endosc. 2017;86:1022–7.e1. [DOI] [PubMed] [Google Scholar]
- [12].Wu KH, Shih HA, Hung MS, Hsiao CT, Chen YC. The association between blood urea nitrogen to creatinine ratio and mortality in patients with upper gastrointestinal bleeding. Arab J Gastroenterol. 2018;19:143–7. [DOI] [PubMed] [Google Scholar]
- [13].Ernst AA, Haynes ML, Nick TG, Weiss SJ. Usefulness of the blood urea nitrogen/creatinine ratio in gastrointestinal bleeding. Am J Emerg Med. 1999;17:70–2. [DOI] [PubMed] [Google Scholar]
- [14].Machlab S, Garcia-Iglesias P, Martinez-Bauer E, Campo R, Calvet X, Brullet E. Diagnostic utility of nasogastric tube aspiration and the ratio of blood urea nitrogen to creatinine for distinguishing upper and lower gastrointestinal tract bleeding. Emergencias. 2018;30:419–23. [PubMed] [Google Scholar]
- [15].Zia Ziabari SM, Rimaz S, Shafaghi A, et al. Blood urea nitrogen to creatinine ratio in differentiation of upper and lower gastrointestinal bleedings; a diagnostic accuracy study. Arch Acad Emerg Med. 2019;7:e30. [PMC free article] [PubMed] [Google Scholar]
- [16].Tielleman T, Bujanda D, Cryer B. Epidemiology and risk factors for upper gastrointestinal bleeding. Gastrointest Endosc Clin N Am. 2015;25:415–28. [DOI] [PubMed] [Google Scholar]
- [17].Anglin R, Yuan Y, Moayyedi P, Tse F, Armstrong D, Leontiadis GI. Risk of upper gastrointestinal bleeding with selective serotonin reuptake inhibitors with or without concurrent nonsteroidal anti-inflammatory use: a systematic review and meta-analysis. Am J Gastroenterol. 2014;109:811–9. [DOI] [PubMed] [Google Scholar]
- [18].Andrade C, Sharma E. Serotonin reuptake inhibitors and risk of abnormal bleeding. Psychiatr Clin North Am. 2016;39:413–26. [DOI] [PubMed] [Google Scholar]
- [19].Yuet WC, Derasari D, Sivoravong J, Mason D, Jann M. Selective serotonin reuptake inhibitor use and risk of gastrointestinal and intracranial bleeding. J Am Osteopath Assoc. 2019;119:102–11. [DOI] [PubMed] [Google Scholar]
- [20].He Y, Chan EW, Leung WK, Anand S, Wong IC. Systematic review with meta-analysis: the association between the use of calcium channel blockers and gastrointestinal bleeding. Aliment Pharmacol Ther. 2015;41:1246–55. [DOI] [PubMed] [Google Scholar]
- [21].Nishiguchi MA, Spencer CA, Leung DH, Leung TH. Aging suppresses skin-derived circulating SDF1 to promote full-thickness tissue regeneration. Cell Rep. 2018;24:3383–92.e5. [DOI] [PMC free article] [PubMed] [Google Scholar]