In-hospital mortality and organ failure after open and endovascular thoraco-abdominal aortic surgery can be predicted by increased levels of circulating dipeptidyl peptidase 3

Abstract

 

Open in new tabDownload slide

OBJECTIVES

Endovascular and open thoraco-abdominal aortic aneurysm (TAAA) repair is associated with specific complications. Circulating dipeptidyl peptidase 3 (cDPP3) is a novel biomarker that shows a strong association with organ failure which has not been assessed in surgical settings. Therefore, the objective of this study was to assess the prognostic capabilities of cDPP3 for predicting patient survival and organ failure following open and endovascular TAAA repair.

METHODS

Thirty-three patients undergoing TAAA repair were assessed in this prospective observational single-centre study. cDPP3 levels were serially measured perioperatively until 72 h after admission to the intensive care unit (ICU). In-hospital mortality and any organ failure were the clinical end points.

RESULTS

Postoperative organ failure was detected in 17 patients (51.5%), and 6 patients died after surgery (18.2%). At 12 h after admission to the ICU, cDPP3 levels were significantly increased in patients who died or developed organ failure (P < 0.001). cDPP3 levels after surgery demonstrated a remarkable predictive accuracy for in-hospital mortality [12 h area under the receiver operating characteristic curve (AUC): 0.907 (P < 0.001), 24 h AUC: 0.815 (P = 0.016), 48 h AUC: 0.914 (P = 0.003)] and the development of organ failure [12 h AUC: 0.882 (P < 0.001), 24 h AUC: 0.850 (P < 0.001), 48 h AUC: 0.846 (P < 0.001)]. Additionally, a significant correlation between cDPP3, the sequential organ failure assessment score and procalcitonin, C-reactive protein and interleukin-6 levels (P < 0.001, P < 0.001, P = 0.011, P = 0.007, respectively) based on all available measurements and time points was observed.

CONCLUSIONS

The present findings highlight the role of cDPP3 as an early, highly specific postoperative biomarker for prediction of in-hospital mortality and organ failure after TAAA repair.

INTRODUCTION

Patients frequently develop organ dysfunction after thoraco-abdominal aortic aneurysm (TAAA) surgery, which ultimately leads to high morbidity and mortality rates, regardless of whether surgery is performed by open or endovascular means [1, 2].

Extensive ischaemia-reperfusion damage during and after open TAAA triggers systemic inflammatory response syndrome, which often further contributes to the development of multiple organ failure, shock and sepsis [3, 4]. To allow timely treatment for complications, readily accessible specific biomarkers for the detection of adverse outcomes would be useful to improve patient treatment. Recently, circulating dipeptidyl peptidase 3 (cDPP3), an amino dipeptidyl peptidase involved in the degradation of angiotensin II and enkephalins, has been described as a biomarker for in-hospital mortality and short-term outcomes. Critically ill patients with haemodynamic instability suffering from cardiogenic shock showed increased levels of cDPP3 [5–7]. After TAAA repair, patients frequently demonstrate haemodynamic instability, and surgery per se represents an ideal model to study the kinetics and clinical significance of cDPP3 in a clinical model of predictable ischaemia/reperfusion and inflammation. Therefore, we evaluated the clinical relevance of cDPP3 in patients undergoing open and endovascular TAAA repair with a special focus on the development of organ dysfunction and in-hospital mortality.

METHODS

Between January and December 2017, a total of 33 patients were included in this prospective, observational single-centre study upon obtaining written informed consent. The study was approved by the local internal review board (EK004/14). Each patient provided informed consent prior to inclusion in the study, and the study was conducted in accordance with the principles outlined in the Declaration of Helsinki.

Patients were included if an endovascular or open TAAA repair was performed. TAAA was defined according to the Crawford classification [8]. The exclusion criteria were an age below 18 years, pregnancy, chronic kidney disease requiring permanent dialysis treatment, emergency procedures and ongoing immunosuppressive medication therapy.

Data collection

The daily physiological variables, demographics (age, sex, body mass index and medical history including history of cardiovascular disease, diabetes and chronic kidney insufficiency) and type of surgery were determined based on medical records and electronic bedside flow charts (Tables 1 and 2). Before surgery, after admission to the intensive care unit (ICU) and 12, 24, 48 and 72h after admission to the ICU, blood samples were collected.

In this study, all complications have been summarized as ‘any organ failure’. Sepsis was defined according to the guidelines of the German Sepsis Society [9]. Shock was defined as recommended by Bone et al. [10].

Cardiac complications included myocardial infarction, acute heart failure and ventricular tachycardia [11]. Pneumonia and tracheotomy were defined according to the guidelines of the American Thoracic Society or the Belgian Society of Pneumology, respectively [12, 13]. Acute kidney injury within 48 h was defined according to the Kidney Disease Improving Global Outcomes criteria based on serum creatinine levels [14]. The sequential organ failure assessment score, which was developed to assess the acute morbidity due to critical illness, was used [15].

Circulating dipeptidyl peptidase 3 measurement

Human plasma samples were measured using the recently described DPP3 luminescence immunoassay [16].

Surgery

Open and endovascular surgery

Open TAAA repair was performed in physically fit patients depending on their cardiovascular capacity, which was assessed via lung function testing and cardiac ultrasound. Furthermore, the surgeons’ assessment was an additional criterion. In the case of connective tissue disease, such as Marfan syndrome, or in the case of an unsuitable anatomy, such as that associated with complex postdissectional TAAA, open repair was the preferred treatment option. In older, more frail patients suffering from chronic disease or in patients after previous laparotomy of a left thoracotomy, endovascular repair was considered the appropriate treatment.

The operation protocol for open TAAA repair has been previously published in detail [17]. It includes double-lumen endotracheal tube intubation, CSFD, perioperative monitoring of motor evoked potentials, sequential aortic clamping if possible, extracorporeal circulation with distal aortic perfusion and selective visceral perfusion and mild hypothermia at 32–33°C [18]. Custodiol^® (4°C) (Dr Franz Köhler Chemie, Bensheim, Germany) was used for renal perfusion instead of blood perfusion. This procedure has been shown to protect the kidneys from ischaemic organ damage. Thoracolaparotomy through the fifth to eighth intercostal space depending on the extent of the aneurysm was used for surgical access. Postoperatively, the mean arterial pressure was adjusted based on motoric evoked potentials, and the intracranial pressure was kept at ≤10 mmHg during the first 72 h for all patients. Extubation was performed as soon as possible postoperatively.

Endovascular TAAA repair was conducted under general anaesthesia; neuromonitoring was performed identically to the method used for open TAAA repair. The detailed procedure of fenestrated endovascular aortic aneurysm repair has been described before [19]. In the case of endovascular procedure, renal perfusion was not directly interrupted by aortic clamping. Contrast medium was applied to avoid functional impairment of the kidneys, leading to a mean application of 65 ± 17 ml per endovascular procedure.

Statistical analysis

The primary end point was to evaluate the clinical significance of the cDPP3 levels for the prediction of organ failure and in-hospital mortality following open and endovascular aortic repair. This was an exploratory study in TAAA patients that we were able to recruit in a representative 12-month period (2017), to study new biomarkers before and after TAAA surgery. Sample size was considered sufficient for a first, exploratory analysis of serial data in this indication.

Values are expressed as the means and standard deviations, medians and interquartile ranges or counts and percentages, as appropriate. Group comparisons of continuous variables were performed using the Kruskal–Wallis test. Categorical data were compared using Pearson’s χ² test for count data. Receiver operating characteristic curves were constructed to assess the sensitivity and specificity of the cDPP3 concentration obtained at each time point and to compare the ability to predict organ failure or in-hospital mortality. The cDPP3 values were not normally distributed and were therefore log-transformed.

Logistic regression was used to evaluate the association between cDPP3 and the end point in-hospital mortality and organ failure. The area under the receiver operating characteristic curve (AUC) was given as a measure of effect size.

All statistical tests were 2-tailed, and a two-sided P-value of 0.05 was considered significant. P-values were not adjusted for multiple testing. The statistical analyses were performed using R version 3.4.3 (http://www.r-project.org, library rms, Hmisc, ROCR) and the Statistical Package for the Social Sciences version 22.0 (SPSS Inc., Chicago, IL, USA).

RESULTS

Patient characteristics and intraoperative and postoperative data

Thirty-three patients were enrolled over 12 months and followed up until hospital discharge. The mean patient age was 63 (±16.2) years, and 16 patients (48.5%) were women. Endovascular TAAA repair was conducted in 19 patients (57.6%), and open repair was conducted in 14 patients (42.4%). (Table 1).

Postoperative organ failure was observed in 17 patients (51.5%). Acute kidney injury occurred in 17 patients (51.5%), pneumonia occurred in 7 patients (21.2%) and 4 patients (12.1%) required tracheotomy because of weaning failure. Spinal cord ischaemia was observed in 3 patients. One patient suffered from paraplegia after open type II repair, and 2 patients suffered from paraparesis after open type III and endovascular TAAA type II repair. Cardiac complications were found in 10 patients (30.3%) (1 case of myocardial infarction, 4 cases of cardiogenic shock caused by acute heart failure and 5 cases of ventricular tachycardia). Sepsis was found in 2 patients (6.1%). The in-hospital mortality rate was 18.2% (n = 6) (Table 2).

Time course of cDPP3 levels following open and endovascular aortic aneurysm repair

The perioperative time course of the cDPP3 levels showed an increase immediately on admission to the ICU, peaking as early as 12 h post admission, followed by normalization at 72 h post-surgery (P < 0.001). Patients with endovascular and open TAAA repair showed similar dynamics and characteristics, with the remarkable exception that cDPP3 levels peaked earlier in patients with open TAAA repair {cDPP3 level for open vs endovascular TAAA immediately after surgery: AUC 0.91 [confidence interval (CI) 0.81–1.0), P < 0.001; P > 0.2 for all other time points} (Fig. 1).

Clinical significance of the perioperative cDPP3 levels

No significant correlation between preoperative clinical variables and baseline cDPP3 levels could be observed. Significant correlations were found between postoperative cDPP3 levels and the postoperative sequential organ failure assessment score, procalcitonin levels and the blood levels of c-reactive protein, interleukin-6 and serum creatinine (P < 0.001, P < 0.001, P = 0.011, P = 0.007, P = 0.024, respectively) at all measured time points (Supplementary Material, Tables S1 and S2). Intraoperative and postoperative mass transfusion had a moderate impact on DPP3 levels at ICU admission, with a correlation between the level of cDPP3 and the number of blood transfusion units of r = 0.52 (CI 0.21–0.72, P = 0.003). In a logistic regression model, correcting the cDPP3 levels by the number of blood transfusion units did not substantially change the odds ratio (OR) for cDPP3 at admission [univariate OR 58 (CI 5–720), adjusted OR 51 (CI 4–651)]. Therefore, the observed differences in cDPP3 levels between patients with open and endovascular surgery at ICU admission cannot be explained by the number of blood transfusion units.

Predictive accuracy of cDPP3 for the development of postoperative organ dysfunction and mortality

At an early postoperative time point at 12 h after ICU admission, cDPP3 levels were significantly increased in patients who died or developed organ failure (P < 0.001) during the ICU stay. At this time point, elevated blood levels of cDPP3 were highly and significantly predictive of the development of organ failure [12 h AUC: 0.882 (P < 0.001), 24 h AUC: 0.850 (P < 0.001), 48 h AUC: 0.846 (P < 0.001)] and in-hospital mortality [12 h AUC: 0.907 (P < 0.001), 24 h AUC: 0.815 (P = 0.016), 48 h AUC: 0.914 (P = 0.003)]. All details can be found in Figs 2A and B and 3A and B as well as Table 3.

DISCUSSION

Patients undergoing open and endovascular TAAA repair frequently show a pertinent rate of severe postoperative complications leading to organ failure and in-hospital mortality, most frequently, acute kidney injury and pneumonia [20, 21]. In this scenario, the use of biomarkers for risk assessment may have a positive impact on early patient treatment.

DPP3 is an intracellular amino dipeptidyl peptidase that cleaves dipeptides from the N-termini of bioactive substrates [5]. In addition, Menale et al. [22] showed in a mouse DPP3-knockout model that a lack of DPP3 results in sustained oxidative stress.

The known substrates of DPP3 are extracellular peptides and include angiotensins, endorphins and enkephalins [5]. Intravenous injections of DPP3 induce myocardial depression, suggesting that angiotensin metabolism might be affected by high DPP3 blood levels [7]. Furthermore, high cDPP3 blood levels at admission in cardiogenic shock patients were associated with severe organ dysfunction, refractory shock and high short-term mortality, while reduction of cDPP3 levels within 24 h of admission was associated with an improved outcome [7, 23, 24].

The current hypothesis suggests that DPP3 is released into the bloodstream upon massive cell death as occurs during invasive surgeries that are followed by systemic inflammation. However, the time course of cDPP3 levels and its clinical significance in terms of postoperative outcomes, especially after TAAA repair have not yet been addressed. In this respect, this is the first study assessing perioperative cDPP3 levels and their significance in terms of patient outcomes.cDPP3 levels were characterized by a rapid postoperative increase, which was followed by a decrease until 72 h after surgery. Notably, postoperatively elevated cDPP3 levels were closely associated with poor outcomes, as assessed by the patient sequential organ failure assessment score after surgery. The significant correlation between cDPP3 levels and the levels of procalcitonin and IL-6 further supports the relevance of cDPP3 as a potential biomarker in this setting that is associated with the extent of surgery-related inflammation and the severity of the underlying disease [25].

The present findings further demonstrate that elevated cDPP3 levels, as measured early postoperatively after ICU admission, showed remarkable predictive accuracy for the development of organ failure and ultimately in-hospital mortality. The specific AUC values reported in this study underline the potential role of postoperatively assessed cDPP3 levels as a promising predictive biomarker for the detection of organ dysfunction in patients undergoing either open or endovascular TAAA repair, which are both associated with an increased risk of severe complications [26].

The prediction of adverse outcomes after these extensive surgical procedures is difficult but of clinical relevance. The postoperative course after major surgical procedures, such as TAAA, is characterized by the overwhelming release of markers of inflammation and is associated with complications during the first 48 h after surgery. It is, therefore, extremely challenging to identify useful biomarkers in this time window. In this context, the use of biomarkers for risk assessment and outcome prediction for in-hospital mortality and organ failure would be a desirable tool for the adequate initiation of treatment bundles and may support the decision-making process leading to improvement in patient outcomes. The current availability of DPP3 detection as a point-of-care test could be part of a standard intra- or postoperative testing system including several established and new biomarkers used to start postoperative therapy in a timely and accurate manner.

In the context of TAAA surgery, urinary neutrophil gelatinase-associated lipocalin and bioadrenomedullin (bio-ADM) have been described as potential biomarkers of adverse outcomes [27]. In general, the identification of new blood-based biomarkers might enable their integration into postoperative surveillance, leading to an improved diagnostic routine.

While comparing the findings of this study with those of trials focusing on critically ill patients, similar results were observed [7, 23].

Limitations

Relevant limitations of this prospective observational study should be acknowledged for proper interpretation of the results. Only a small number of patients presenting with an inhomogeneous comorbidity profile were enrolled due to the overall low number of TAAA procedures performed yearly worldwide. Even if the results of our study are promising and the test quality is appropriate, the hypothesis-generating character of this study must be emphasized. Especially, the comparison of open and endovascular TAAA repair is, at best, putative and not conclusive, given the size of the database and the uncontrolled nature of the analysis.

The category ‘any organ failure’ is rather non-specific and does not enable the use of specific measures to prevent organ failure; yet, such markers could be used as predictive surrogate parameters for patient risk assessment and outcome prediction. For validation, further clinical studies are required to verify the clinical relevance of cDPP3 as a biomarker for the early detection of in-hospital mortality and organ failure specifically after open or endovascular TAAA surgery. Moreover, a clinically more reliable conclusion focusing on a homogeneous surgical procedure, such as endovascular TAAA repair, would call for a more specific methodology. Furthermore, it could have been appropriate to assess intraoperative DPP3 levels, which could have led to a timelier detection of patients at risk.

CONCLUSION

The present findings demonstrate for the first time the perioperative kinetics of the novel cDPP3 biomarker and show early postoperative cDPP3 levels to be predictive, with high suitability, for the development of organ dysfunction and in-hospital mortality after open and endovascular TAAA repair. If confirmed in other settings of major cardiovascular surgical procedures, the measurement of cDPP3 could improve the current clinical risk prediction models and thus be a helpful decision-making tool for clinicians.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

Funding

SphingoTec has organized the shipment and 4Teen4 has organized the DPP3 measurements at no cost for the Department of Vascular Surgery, RWTH University of Aachen.

Conflict of interest: Oliver Hartmann, Deborah Bergmann and Janin Schulte are employees of SphingoTec. Janin Schulte is an employee of 4TEEN4 Pharmaceuticals GmbH. 4TEEN4 Pharmaceuticals GmbH has patent rights for the DPP3 biomarker. SphingoTec has organized the shipment and 4TEEN4 has organized the DPP3 measurements at no cost for the Department of Vascular Surgery, RWTH University of Aachen. All measurements were performed in a blinded fashion. All other authors declare no conflict of interest.

Author contributions

Alexander Gombert: Data curation; Formal analysis; Project administration; Supervision; Writing—original draft; Writing—review & editing. Mohammad Barbati: Data curation; Writing—original draft. Drosos Kotelis: Writing—review & editing. Tim-Philipp Simon: Conceptualization. Thomas Breuer: Writing—review & editing. Oliver Hartmann: Data curation; Visualization; Writing—original draft; Writing—review & editing. Karine Santos: Writing—review & editing. Deborah Bergmann: Conceptualization; Writing—review & editing. Janin Schulte: Investigation; Writing—review & editing. Gernot Marx: Conceptualization; Project administration. Michael Jacobs: Writing—review & editing. Christian Stoppe: Conceptualization; Writing—review & editing.

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks George J. Arnaoutakis, Luca Di Marco, Roman Gottardi and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.

REFERENCES

Abbreviations

 
• AUC

  Area under the receiver operating characteristic curve

 
• cDPP3

  Circulating dipeptidyl peptidase 3

 
• CI

  Confidence interval

 
• ICU

  Intensive care unit

 
• OR

  Odds ratio

 
• TAAA

  Thoraco-abdominal aortic aneurysm