{
  "abstract": "Introduction Diabetic ketoacidosis (DKA) occurs frequently in children with type 1 diabetes (T1D). The inflammatory response to DKA may play a role in complications, but the inflammatory pattern is not well characterized. We aimed to describe the inflammatory profile during and after DKA.Research design and methods We evaluated inflammatory mediators (cytokines, chemokines, growth factors, and matrix metalloproteinases) using multiplex immunoassays in children (1) hospitalized with acute DKA (6–8 hours after beginning treatment, (n=15), (2) seen in the outpatient diabetes clinic 2–5 days after DKA (n=14), (3) hospitalized with new-onset T1D without DKA <24 hours after beginning insulin (n=9), and (4) referred to the outpatient diabetes clinic for new-onset T1D without DKA 2–5 days after beginning insulin (n=14). Children with chronic T1D and glycated hemoglobin <8.0% (n=59) undergoing routine phlebotomy served as a reference group.Results Compared with the reference group, children with acute DKA had significant alterations in interleukin 1 (IL-1) receptor antagonist (IL-1RA), IL-6, IL-8, IL-10, IL-18, chemokine C-X-C motif ligand (CXCL) 5, CXCL10, chemokine C-C motif ligand (CCL) 27, tumor necrosis factor-related apoptosis-inducing ligand, granulocyte colony-stimulating factor, tissue inhibitor of metalloproteinase 2 (TIMP-2), TIMP-4, matrix metalloproteinase 2 (MMP-2), MMP-3, MMP-7, MMP-9, and MMP-10. MMP-3, MMP-10, TIMP-1, and IL-1RA were also elevated 2–5 days after DKA (false discovery rate-adjusted p<0.10 for all). MMP-2 and MMP-9 levels were altered in children with new-onset T1D without DKA <24 hours after starting insulin, but no significant inflammatory changes were found in new-onset T1D 2–5 days after starting insulin.Conclusions DKA causes a unique inflammatory pattern distinct from inflammatory changes in acute hyperglycemia or T1D-related autoimmunity. Alterations in MMPs and their tissue inhibitors play a dominant role in this inflammatory profile.",
  "authors": [
    {
      "affiliations": [
        "Department of Pediatrics, University of California Davis School of Medicine, Sacramento, California, USA"
      ],
      "name": "Zachary Chaffin"
    },
    {
      "affiliations": [
        "Department of Psychology, University of California Davis, Davis, California, USA"
      ],
      "name": "Simona Ghetti"
    },
    {
      "affiliations": [
        "Department of Pediatrics, University of California Davis School of Medicine, Sacramento, California, USA"
      ],
      "name": "Daniel Tancredi"
    },
    {
      "affiliations": [
        "Department of Pediatrics, The Colorado Children’s Hospital, University of Colorado Denver Anschutz Medical Campus, Aurora, Colorado, USA"
      ],
      "name": "Arleta Rewers"
    },
    {
      "affiliations": [
        "Department of Pediatrics, The Colorado Children’s Hospital, University of Colorado Denver Anschutz Medical Campus, Aurora, Colorado, USA"
      ],
      "name": "Marian Rewers"
    },
    {
      "affiliations": [
        "Department of Pediatrics, University of California Davis School of Medicine, Sacramento, California, USA"
      ],
      "name": "Spencer Gilles"
    },
    {
      "affiliations": [
        "Department of Pediatrics, University of California Davis School of Medicine, Sacramento, California, USA"
      ],
      "name": "Bradley Ander"
    },
    {
      "affiliations": [
        "Department of Pediatrics, University of California Davis School of Medicine, Sacramento, California, USA"
      ],
      "name": "Nicole Glaser"
    }
  ],
  "full_text": "WHAT IS ALREADY KNOWN ON THIS TOPIC Previous studies evaluated levels of select inflammatory mediators during diabetic ketoacidosis (DKA), but no previous study has comprehensively described the inflammatory response to DKA.WHAT THIS STUDY ADDS The inflammatory response to DKA involves acute and delayed phases. Alterations in matrix metalloproteinases and their tissue inhibitors dominate both phases of this response.HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY Our findings suggest that investigation of the role of matrix metalloproteinases in causing complications of DKA is warranted.Introduction Diabetic ketoacidosis (DKA) is a common and potentially life-threatening complication of type 1 diabetes (T1D) in children and adolescents. The presence of DKA at the initial presentation of T1D has been increasing in frequency over the past two decades. 1–3 DKA is an important cause of morbidity and mortality in children with T1D, with both acute and chronic manifestations. Although complications of DKA affecting several organs have been well documented, the causes of these remain poorly understood. The most feared acute complication of DKA in children is cerebral injury, the primary cause of DKA-related mortality. Although life-threatening cerebral injury is rare, subtle injuries to the brain are common during DKA and can have long-term effects on cognition.4–6 Acute kidney injury (AKI) is another increasingly recognized complication of DKA, occurring in 40%–60% of pediatric DKA episodes.7–10 Recent studies demonstrate that the occurrence of AKI during DKA is associated with increased frequency of development of albuminuria, a risk marker for chronic diabetic kidney disease.11 Notably, children who developed AKI during DKA performed more poorly on cognitive tests after DKA resolution compared with children with similar severity of DKA but without AKI.10 These findings suggest pathophysiologic concurrence of injuries to the brain and kidneys during DKA and raise the question of a common underlying systemic response triggered by DKA affecting multiple organs.Previous studies have documented alterations in various proinflammatory cytokines and chemokines during DKA.12–14 Alterations in matrix metalloproteinase levels have also been reported in DKA and have been hypothesized to damage blood–brain barrier integrity, contributing to cerebral edema.15 16 Although previous studies have measured select groups of inflammatory mediators, no previous study has comprehensively described the inflammatory profile during DKA. We hypothesized that there is a unique, systemic inflammatory response triggered by DKA distinct from that caused by hyperglycemia or autoimmunity at the onset of T1D. The goal of the current study was to characterize the inflammatory profile of DKA in children and adolescents.Methods We prospectively enrolled children with T1D between February 2021 and December 2022 at both participating hospitals. Children admitted with DKA who were undergoing phlebotomy according to the hospital DKA protocol or children attending diabetes clinic visits during which phlebotomy was planned for routine care were approached for enrollment. The pediatric critical care physician group and the pediatric endocrinology physician group were advised of the study and asked to notify the research coordinator when (1) children were admitted to the critical care unit for DKA or to the pediatric ward for new-onset diabetes without DKA, (2) children with episodes of DKA were scheduled to be seen in the outpatient diabetes clinic within 5 days of DKA treatment, (3) children with new-onset T1D without DKA were scheduled to be seen in the outpatient diabetes clinic within 5 days of starting insulin treatment, or (4) children with T1D seen in the outpatient diabetes clinic would be undergoing phlebotomy for routine diabetes care. Children were approached during times when a research coordinator or one of the study investigators was available to obtain consent and process blood samples. Data were not recorded for children who were not approached or those who declined to participate.After obtaining informed consent from parents/guardians and participant assent for children 7 years or older, we collected blood samples from children in several groups: (1) acute DKA, 6–8 hours after beginning treatment with insulin and intravenous fluids (n=15), (2) DKA recovery (samples collected 2–5 days after DKA resolution (n=14), (3) new-onset T1D without DKA within 24 hours of initiating insulin treatment (n=9), and (4) new-onset T1D without DKA 2–5 days after initiating insulin treatment (n=14). Inflammatory mediator profiles in these groups were compared with a reference group of children with chronic T1D who had glycemic levels at or near target (glycated hemoglobin (HbA1c) less than 8.0% (64 mmol/mol) and were undergoing phlebotomy for routine care (n=59). Study groups were comprised of distinct participants without overlap among groups.Children were eligible for participation if they were diagnosed with T1D and had no other known autoimmune or inflammatory diseases. Children were excluded if they had a fever, other signs of infection, or had been diagnosed with an infectious illness within the 2 weeks preceding enrollment. Children taking medications during the time period under study that could alter inflammatory mediator levels (immune-modifying agents, antibiotics, glucocorticoids) were also excluded. Children were considered to have DKA if they had blood glucose >200 mg/dL (11.1 mmol/L), venous pH <7.25 or serum bicarbonate <18 mmol/L and had a positive urine or blood test for ketones.Blood samples were collected in 5 mL plasma EDTA tubes and gently inverted 10–20 times. Tubes were then centrifuged at 2800 revolutions per minute for 15 min. Plasma samples were divided into five aliquot tubes and stored at −80°C until the time of shipping. For all participants, we assessed a comprehensive panel of 84 inflammatory mediators, including cytokines, chemokines, growth factors, matrix metalloproteinases (MMPs), and tissue inhibitors of matrix metalloproteinases (TIMPs). Multiplex immunoassays were employed for these inflammatory marker measurements using Luminex xMAP technology on the LuminexTM 200 system (Luminex, Austin, Texas, USA) and run by a CLIA-certified lab (Eve Technologies Corp, Calgary, Alberta, Canada). Cytokines, chemokines, and growth factors were simultaneously measured in the samples using several multiplex kits (HCYTA-60K-PX48 and HCP2MAG-62K-PX23 from Millipore Sigma, Burlington, Massachusetts, USA, and #FCSTM07/MMP and #LKTM003/TIMP from R&D Systems, Minneapolis, Minnesota, USA). Duplicate samples were processed for each subject. All assays were run according to the manufacturer’s protocol, with separate runs containing a consistent reference sample to monitor and use for adjusting any technical drift.Inflammatory mediator levels were natural log-transformed to approximate a normal distribution. To account for possible variability among assays (batch effects) as well as for ease of comparison among inflammatory mediators, we calculated batch and group-specific robust Z-scores for each mediator, subtracting the median values from children with glycemic levels at or near target (reference group) to generate a deviation and then dividing these deviations by a robust measure of scale based on the residuals from a regression model of these deviations that include fixed effects for study group and batch. All inflammatory mediator levels were natural log transformed prior to computing robust Z-scores, and the resulting Z-scores were trimmed to lie between −4 and 4 to lessen the influence of skewed outliers. Mean robust Z-scores for each group were compared with the mean in the reference group using a linear regression model. Important covariates that might affect inflammatory marker levels (age, sex, duration of diabetes, and body mass index (BMI)) were included in the regression analyses, as well as a term indicating assay batch.To assess possible confounding effects of new-onset versus previously diagnosed T1D within the DKA group, we performed subanalyses including only children with new-onset T1D in the regression models. To account for repeated comparisons, we used the two-stage Benjamini, Yekutieli and Krieger approach to compute adjusted p-values to control the false discovery rate at 10% for the family of regression-adjusted between-group comparisons pooled across all of the mediators examined as outcomes.17 Demographic measures in each group were compared with the reference group using Student’s t-test for continuous measures and Fisher’s exact test for categorical measures.In exploratory analyses investigating possible associations between inflammatory mediators and DKA complications, we compared MMP and TIMP levels in children with and without AKI during DKA. MMP/TIMP levels were chosen for these comparisons due to the predominance of MMP/TIMP alterations in the inflammatory profile characterizing DKA. We calculated an MMP composite score as the mean of Z-scores for all MMPs and TIMPs that were significantly altered during acute DKA. Z-scores that were inversely associated with DKA were included in the equation as the negative of the calculated Z-score. The presence of AKI during DKA was determined according to the Kidney Disease/Improving Global Outcomes (KDIGO) serum creatinine criteria.18 Because baseline creatinine values were not routinely available for study participants, an estimated glomerular filtration rate of 120 mL/min/1.73m2 was used to calculate the expected baseline creatinine level for each participant using the Schwartz estimating equation.19 Children with creatine levels higher than 1.5 times the estimated baseline creatinine level (stage 1 AKI as defined by the KDIGO criteria18) were considered to have AKI. MMP composite Z-scores in children with and without AKI during DKA were compared using the Wilcoxon rank-sum test. All statistical analyses were performed using STATA V.17.0.Data and resource availability The datasets generated and analyzed during the current study are not publicly available due to ongoing subanalyses and other work using these datasets. However, the datasets are available from the corresponding author upon reasonable request.Results 123 children enrolled in the study. Blood samples could not be collected from six enrolled children (the quantity of blood samples was insufficient to allow for both clinical tests and study labs, or the timing of clinical labs did not coincide with the study sample window). Blood samples from seven children were excluded due to infection recognized after enrollment (n=6) or laboratory error (n=1). The remaining 110 children comprised the study sample ( table 1). Participants in the reference group (children with previously diagnosed T1D with HbA1c <8.0% (64 mmol/mol)) were older than those in the DKA and new-onset T1D groups and had higher BMI. The distributions of sex and race were not significantly different among the comparison groups. The duration of diabetes in the reference group was 87±45.7 months. Patients in the acute DKA group with previously diagnosed diabetes had a duration of 51±25.5 months (p=0.027 compared with the reference group). The DKA recovery group was comprised entirely of children with new-onset T1D, compared with 36% in the acute DKA group (p=0.005). Severity of DKA was similar between the acute DKA group and the DKA recovery group (pH 7.02±0.17 vs 7.03±0.17, p=0.83).Table 1Demographic characteristics of study participantsReference group*:n=59Acute DKA†n=14DKA recovery‡n=14New-onset T1D§n=9New-onset T1D post-treatment¶n=14Age (years)14.5 (2.9)9.9 (3.1)(p<0.001)9.47 (2.0)(p<0.001)9.89 (3.06)(p<0.001)9.4 (2.4)(p<0.001)Sex (% female)53%78%(p=0.38)21%(p=0.08)22%(p=0.28)53%(p=1.0)Body mass index22.7 (4.6)20.9 (4.7)(p=0.23)16.7 (2.4)(p<0.001)19.1 (8.1)(p=0.07)17.7 (3.8)(p<0.001)Race (%)(p=0.08)(p=0.24)(p=1.0)(p=0.04) White86.4%64%79%100%67% Black0%0%0%0%13% Asian1.7%7%0%0%0% Other10.2%0%7%0%13% Unknown1.7%0%14%0%7%*Children with previously diagnosed T1D with glycated hemoglobin <8.0%.†6–8 hours after starting insulin and intravenous fluids.‡2–5 days after insulin and intravenous fluids.§<24 hours after starting insulin treatment.¶2–5 days after starting insulin.DKA, diabetic ketoacidosis; T1D, type 1 diabetes.Inflammatory mediators that were significantly different from the reference group after correcting for false discovery rate are shown in figure 1. During acute DKA, we found significant alterations in interleukin-1 (IL-1) receptor antagonist (IL-1RA), IL-6, IL-8, IL-10, IL-18, CXCL5, CXCL10, CCL27, tumor necrosis factor-related apoptosis-inducing ligand (TRAIL), granulocyte colony-stimulating factor (G-CSF), TIMP-2, TIMP-4, MMP-2, MMP-3, MMP-7, MMP-9, and MMP-10 (figure 1a).Figure 1(a) Z-score boxplots for inflammatory mediators that were significantly altered in children with acute DKA (false discovery rate adjusted p<0.10). (b) Inflammatory mediator Z-score boxplots in children 2–5 days after DKA recovery. Inflammatory mediators that were significantly different from the reference group are shown in dark gray. Those not significantly different from the reference group are shown in light gray. (c) Inflammatory mediator Z-score boxplots in children with new-onset T1D. Inflammatory mediators that were significantly different from the reference group are shown in dark gray. Those not significantly different from the reference group are shown in light gray. The following inflammatory mediators were not significantly altered by DKA or new-onset T1D (data not shown): sCD40L, FGF2, FLT3L, CX3CL1, IFNα2, IFNγ, IL-1α, IL-1β, IL-2, IL-3, IL-4, IL-5, IL-7, IL-9, IL12p40, IL12p70, IL-13, IL-15, IL-16, IL-17A, IL-17E, IL-17F, IL-20, IL-21, IL-22, IL-23, IL-27, IL-28A, IL-33, CCL1, CCL2, CCL3, CCL4, CCL5, CCL7, CCL8, CCL11, CCL12, CCL13, CCL17, CCL21, CCL22, CCL24, CCL26, CXCL9, CXCL13, MCSF, PDGFAA, PDFABBB, TGFα, TNFα, TNFβ, VEGF-A, LIF, MIP-1δ, SCF, TPO, TSLP, MMP-1, MMP-12.CCL, chemokine C-C motif ligand, CXCL, chemokine C-X-C motif ligand, CX3CL, chemokine C-X3-C motif ligand; DKA, diabetic ketoacidosis; G-CSF, granulocyte colony-stimulating factor; FGF2, fibroblast growth factor 2; FLT3L, FLT3 ligand; IFNα2, interferon alpha 2; IFNγ, interferon gamma; IL-1RA, interleukin-1 receptor antagonist; LIF, leukemia inhibitory factor; MCSF, macrophage colony-stimulating factor; MIP-1δ, macrophage inflammatory protein-1δ; MMP, matrix metalloproteinases; PDGF, platelet-derived growth factors; sCD40L, soluble CD40L; SCF, stem cell factor; T1D, type 1 diabetes; TIMPs, tissue inhibitors of matrix metalloproteinases; TNFα, tumor necrosis factor alpha; TNFβ, TNF beta; TPO, thrombopoietin; TSLP, thymic stromal lymphopoietin; TRAIL, tumor necrosis factor-related apoptosis-inducing ligand; VEGF-A, vascular endothelial growth factor A.MMP-3, MMP-10, TIMP-1, and IL-1RA were also elevated 2–5 days after recovery from DKA (figure 1b). MMP-2 and MMP-9 levels were significantly altered in children with new-onset T1D without DKA when measured within 24 hours of starting insulin treatment (figure 1c); however, no inflammatory mediators were significantly altered in the new-onset group several days after starting insulin treatment (data not shown).Because proportions of children with new-onset versus established diabetes differed in the acute DKA group compared with the DKA recovery group, we conducted a subanalysis including only children with new-onset diabetes in both groups. We compared these groups to the same reference group as the main analysis (online supplemental table 1). Although the statistical power to detect differences was reduced in this analysis involving a smaller number of patients, the results were similar. Most mediators identified in the main analysis as associated with acute DKA (IL-6, IL-18, CCL27, TRAIL, MMP-2, MMP-9, TIMP-2, TIMP-4, MMP-10, and MMP-3) continued to be significantly associated in the subanalysis. Others (G-CSF, IL-10, MMP-7, IL-8, CXCL10, and CXCL5) had similar regression coefficients to the larger model; however, these associations did not reach the level of significance. Results of the subanalysis were substantially different for only one of the inflammatory mediators that was associated with acute DKA in the main analysis (IL-1RA).SP110.1136/bmjdrc-2025-004961.supp1Supplementary dataBecause the mean age of the reference group was older than that of the comparison groups, we performed an additional sub-analysis using a smaller reference group that was matched in age with the acute DKA group (age range 6–16 years, n=43). Mean age was no longer significantly different between the acute DKA and reference groups (12.6±3.5 vs 13.4±2.6, p=0.35), although it remained significantly different in the new-onset T1D and DKA recovery groups compared with the reference group. The results of this subanalysis were almost identical to the original analysis with respect to differences in inflammatory mediators between groups (online supplemental table 2).Finally, in exploratory analyses, we compared MMP/TIMP composite scores (mean of Z-scores for MMPs/TIMPs that were significantly altered during acute DKA) in children with AKI during DKA (n=8) to those without AKI during DKA (n=6). We found that MMP composite scores were significantly higher in children with AKI during DKA compared with those without this complication (p=0.03, figure 2).Figure 2MMP/TIMP composite scores in children with and without AKI during DKA. Composite scores represent the mean of Z-scores for all MMPs and TIMPs that were significantly altered during acute DKA. Z-scores that were inversely associated with DKA were included in the equation as the negative of the calculated Z-score; p=0.03 for comparison of children with and without AKI during DKA. AKI, acute kidney injury; DKA, diabetic ketoacidosis; MMP, TIMPDiscussion Our findings suggest that DKA is associated with a unique pattern of alterations in inflammatory mediators that are distinct from inflammatory changes caused by acute hyperglycemia and T1D-related autoimmunity. Alterations in MMPs and TIMPs play a dominant role in the inflammatory profile of DKA. To our knowledge, this is the first comprehensive description of inflammatory mediator profiles during and after DKA in children with T1D. Understanding the inflammatory pattern triggered by DKA may help to guide further investigation of the mechanisms of acute and chronic organ injuries resulting from DKA, particularly the role played by MMPs in causing these injuries.The observed changes in inflammatory mediators during DKA are primarily proinflammatory in the acute phase and would be expected to contribute to the activation of multiple immune cell lineages, including macrophages (IL-6, IL-8, and IL-18) and neutrophils (CXCL5 and IL-8). Of note, we also found changes that are likely immunomodulatory in nature, such as increased levels of IL-10 and IL-1RA, which have been reported to counterbalance the effects of other proinflammatory signals.20 IL-1RA was one of the inflammatory signals that persisted for several days after treatment of DKA, suggesting ongoing immune modulation. Of note, alterations in IL-1RA levels were not significant in a subanalysis including only children with new-onset diabetes in the acute DKA group. These results raise the possibility that there may be differences in the inflammatory profile during DKA for children with new-onset versus long-standing diabetes.Our findings are consistent with the results of prior studies evaluating small groups of inflammatory mediators in children during DKA treatment: increased levels of IL-6, IL-8, and IL-10 have all been previously reported during acute DKA.21–23Our finding of decreased levels of CXCL10 during acute DKA is also consistent with prior work.21 CXCL10 induces chemotactic activity for multiple types of immune cells, such that its downregulation during acute DKA may reflect an immune modulatory process. In contrast to some previous studies, we did not find significant differences in levels of TNF-alpha or IFN-gamma during DKA.13 22 24 Lack of significant differences in levels of these cytokines may have reflected the modest sample size in the DKA group and relatively conservative statistical procedures employed to correct for multiple comparisons.Our findings expand on prior reports of alterations in levels of MMPs and their endogenous tissue inhibitors (TIMPs) during DKA. Levels of MMP-2, MMP-8, MMP-9, and TIMP-4 have been reported to be altered in children with acute DKA,15 16 and these changes have been shown to correlate with DKA severity.16 In the current study, we found that the majority of MMP/TIMPs (8 of 13 MMP/TIMPs measured) were significantly altered during DKA. Furthermore, elevated MMP-3 and MMP-10 levels persisted for several days after DKA recovery. Alterations in MMPs are important because MMP-induced reductions in blood-brain barrier integrity may enhance neural-immune interactions between the brain and peripheral immune cells, possibly contributing to brain injury. In diabetes, alterations in circulating MMP concentrations are associated with long-term complications and with all-cause mortality.25–29 MMPs and their tissue inhibitors (TIMPs) are important factors contributing to kidney fibrosis in diabetic kidney disease,30 and altered urine and plasma levels of MMPs and TIMPs are correlated with the progression of diabetic kidney disease in humans and animal models.29–32 Notably, we found that MMP composite scores were significantly higher in children with AKI during DKA compared with those without AKI. These findings suggest that the possible role of altered MMPs in causing DKA-related AKI should be further explored.Our findings suggest that alterations in MMP levels play a substantial role in the inflammatory response triggered by DKA, and their possible involvement in acute and chronic DKA-related organ injuries should be investigated. Pharmacological interventions to modify MMPs during DKA, for example, the use of minocycline, which is a potent inhibitor of MMP-9, may represent important areas for future exploration.33This pilot study is subject to several limitations. First, this was a descriptive analysis, such that the linear model coefficients do not provide a robust comparison among all groups. Instead, we focused on comparing each group to the reference group for the purpose of describing the inflammatory signature of each specific condition in relation to T1D with HbA1c at or near target. Second, the group of children with acute DKA included both children with new-onset T1D as well as children with previously diagnosed T1D. Findings of a sub-analysis involving only children with new-onset T1D were substantially similar to the main analysis, suggesting that inclusion of children with previously diagnosed T1D in the acute DKA group did not substantially affect the results. Nevertheless, future larger studies should establish whether T1D duration alters inflammatory profiles during acute DKA. The current study also did not include measurements of inflammatory markers before beginning DKA treatment. The inflammatory profile of untreated DKA is likely to differ from that found during treatment, particularly given the known anti-inflammatory effects of insulin.34 Future studies comparing inflammatory profiles before DKA treatment versus during treatment would be of interest. In addition, the aim of this pilot study was to determine how DKA alters inflammation in children with T1D, above and beyond inflammatory alterations caused by autoimmunity or other aspects of T1D. We therefore used children with T1D and glycemic levels at or near target as our reference group rather than healthy control children. Characterizing the inflammatory state in T1D in comparison to healthy controls would also be of interest for future studies. Finally, because the study involved multiple comparisons, we employed procedures to control for the false discovery rate. Criteria to consider inflammatory mediators significantly altered in any of the comparison groups were therefore relatively stringent, and some inflammatory mediator alterations that were less dramatic may not have been identified.Conclusions DKA has a unique inflammatory profile with changes in multiple inflammatory mediators, including IL-6, IL-8, IL-10, IL-18, IL-1RA, CXCL5, CXCL10, CCL27, G-CSF, TRAIL, MMP-2, MMP-3, MMP-7, MMP-9, MMP-10, TIMP-1, TIMP-2, and TIMP-4. These changes are primarily proinflammatory with an early, transient effect (during DKA treatment) and a more prolonged effect that is detectable 2–5 days after recovery from DKA. MMPs and TIMPs are heavily involved in the inflammatory response to DKA. New-onset diabetes without DKA (within 24 hours of initiating insulin treatment) involved fewer inflammatory alterations, being characterized only by changes in MMP-2 and MMP-9. These changes were no longer present several days after beginning insulin treatment, suggesting that changes in MMP-2 and MMP-9 are likely stimulated by hyperglycemia or low insulin levels. Further studies are needed to investigate associations between the inflammatory response to DKA and clinical patterns of organ injury.",
  "title": "Inflammatory profile of diabetic ketoacidosis in children with type 1 diabetes",
  "uid": "59e40dec-2059-5c92-87b3-82c514856a24"
}