Perioperative Renal Ultrasonography of Arterio-to-Venous Coupling Predicts Postoperative Complications after Major Laparoscopic Urologic Surgery

Point-of-care ultrasonography (POCUS) with concomitant venous and arterial Doppler assessment enables clinicians to assess organ-specific blood supply. To date, no studies have investigated the usefulness of including a comprehensive perioperative POCUS assessment of patients undergoing major laparoscopic surgery. The primary aim of the present study was to evaluate whether the combined venous and arterial renal flow evaluation, measured at different time points of perioperative period, may represent a clinically useful non-invasive method to predict postoperative acute kidney injury (AKI) after major laparoscopic urologic surgery. The secondary outcome was represented by the development of any postoperative complication at day 7. We included 173 patients, subsequently divided for analysis depending on whether they did (n = 55) or did not (n = 118) develop postoperative AKI or any complications within the first 7 days. The main results of the present study were that: (1) the combination of arterial hypoperfusion and moderate-to-severe venous congestion inferred by POCUS were associated with worst outcomes (respectively, HR: 2.993, 95% CI: 1.522–5.884 and HR: 8.124, 95% CI: 3.542–18, p < 0.001); (2) high intra-operative abdominal pressure represents the only independent determinant of postoperative severe venous congestion (OR: 1.354, 95% CI: 1.017–1.804, p = 0.038); (3) the overall number of complications relies on the balance between arterial inflow and venous outflow in order to ensure the adequacy of peripheral perfusion; and (4) the overall reliability of splanchnic perfusion assessment by Doppler is high with a strong inter-rater reliability (ICC: 0.844, 95% CI: 0.792–0.844). The concomitant assessment of arterial and venous Doppler patterns predicts postoperative complications after major laparoscopic urologic surgery and may be considered a useful ultrasonographic biomarker to stratify vulnerable patients at risk for development of postoperative complications.


Introduction
Perioperative AKI is a serious complication associated with high morbidity and mortality [1], occurring in 30-40% of all major surgical interventions [2,3]. Even after complete recovery from AKI, patients may develop chronic or end-stage renal disease [4], with an up to 12.6-fold relative risk of death [5]. Moreover, not only is mortality higher, but also non-renal post-surgical complications are more frequent in these patients [6,7].
While close attention is generally paid to pre-and postoperative management, the role of surgical techniques and perioperative management in the development of AKI and other complications remains underappreciated.
Postoperative AKI may be due to either sudden arterial hypoperfusion causing acute tubular necrosis, or renal venous congestion determined by intraluminal hypertension. These two mechanisms can lead to a dramatic fall in glomerular filtration rate; in fact, the net ultrafiltration pressure is derived from the opposing forces of the blood hydrostatic pressure in the glomerulus versus the combination of capsular hydrostatic pressure and blood colloid osmotic pressure [8].
Point-of-care ultrasonography (POCUS) with arterial Doppler assessment enables clinicians to assess organ-specific blood supply in both arteries and veins [9,10]. Recently, Doppler patterns of intrarenal venous flow (IRVF) have been described to predict acute kidney injury (AKI) and post-surgical complications after open heart surgery [11]. However, to date, no studies have investigated the usefulness of including IRVF in a perioperative POCUS assessment of patients undergoing major laparoscopic surgery.
Our hypothesis is that not only arterial perfusion but even venous drainage may influence organ perfusion, determining AKI and other postoperative complications. The primary aim of the present study is to evaluate whether, in patients undergoing major laparoscopic urologic surgery, the combination of IRVF and renal Doppler resistive index (RDRI) measured at different time points of perioperative period represents a clinically useful non-invasive method to predict postoperative AKI (within the first seven days) [12]. The secondary aim is to evaluate if the IRVF and RDRI can predict the occurrence of any other postoperative complication at day seven.

Methods
This is an observational, monocentric, prospective study. The study was approved by the local ethics committee (CER Liguria: 7/2019-id: 4378) and all the enrolled patients gave written informed consent to the use of their data.

Patients
Between December 2019 and April 2022, all the adult patients admitted for major laparoscopic surgery to the urology section of the Galliera Hospital, Genova, Italy, were evaluated for enrollment.
Inclusion criteria were: age > 18 aa, the ability to provide informed consent, and availability of the operators to perform a POCUS assessment before surgery.
Patients were excluded if they had one or more of the following: inability to sample Doppler indices, severe chronic kidney disease (estimated glomerular filtration rate < 15 mL/min per 1.73 m 2 or dialysis), renal transplantation (planned or received), critical preoperative state (defined as preoperative mechanical ventilation, vasopressor or inotropic support), emergency surgery, obstructive uropathy, urinary infection, or sepsis (Supplemental Figure S1).

Data Collection
Demographic and baseline clinical data were collected including the Charlson comorbidity index (CCI), the Karnofsky scale, and the preoperative American Society of Anesthesiologists Classification (ASA score). Creatinine measurements were obtained before surgery and daily after surgery until discharge. The estimated glomerular filtration rate was calculated using the Modified Diet in Renal Disease formula [13]. AKI was defined accordingly to the Kidney Disease Improving Global Outcomes criteria, based on creatinine values and urinary output variations [14]. Intraoperative, clinical, and laboratory data were monitored until hospital discharge.

Renal Ultrasound Assessment
All ultrasonographic examinations were performed by two expert operators using a Mindray TE7 ultrasound device (Shenzhen Mindray Bio-Medical Electronics Co., Shenzhen, China) with a 2.5-5.0 MHz convex array transducer for kidney's examination with the patient in a supine position [15,16]. After a general preliminary examination of the abdominal cavity and organs, Doppler ultrasound measurements were obtained along the interlobar vessels of the renal cortex.
The ultrasound examination was considered technically acceptable if the following criteria were met: (a) the possibility to obtain a clear two-dimensional longitudinal scan with definition of renal parenchyma, (b) a good color Doppler image with visualization of IRVF was feasible, and (c) at least three consecutive Doppler time-velocity spectra were obtained. Pulsed-wave Doppler spectrum was increased by using the lowest frequency-shift range that did not cause aliasing, and the wall filter was set at a low frequency (100 MHz). RDRI was calculated according to Planiol and Pourcelot's protocol as (VMax-VMin)/VMax [17]. The value of Doppler measurements for IRVF was calculated and averaged to minimize sampling error. The IRVF patterns were classified as continuous, biphasic, or monophasic, depending on their presence during the whole heart cycle, intermittency with a systolic and a diastolic peak, or the only presence of a diastolic peak [18].
All Doppler examinations were recorded on the day before surgery, at postoperative day 1, and at hospital discharge. The attending physician was unaware of the results of the ultrasound examinations. The investigators were blinded to the clinical parameters while performing the ultrasound assessments.

Statistical Analysis
All the variables are expressed as median and interquartile range (IQR) for continuous variables, or as percentage (%) for categorical variables. The Shapiro-Wilk test was used to evaluate the normal distribution of continuous variables. The Mann-Whitney U-test or Fisher's exact test were used to evaluate differences between groups for continuous or categorical variables, respectively. Cumulative probability for lack of postoperative AKI or any complications in patients with different Doppler profiles were calculated with the Kaplan-Meier product-limit estimator. The censored/uncensored patients corresponded to hospital discharge/occurrence of complications. The log-rank (Mantel-Cox) test was applied to evaluate the difference between Kaplan-Meier curves in Doppler profiles [19].
The main Doppler and clinical variables were entered into a Cox proportional hazard analysis. The variables associated in univariate models with occurrence of postoperative AKI or any complications with a p < 0.10 without violating the assumption of proportional risk were entered into a multivariate Cox proportional hazard model. In each Cox model both the regression coefficient and hazard ratio (HR) with 95% confidence interval (CI) were calculated. Logistic regression was used to identify independent variables potentially associated with IRVF patterns identified as predictors of complications, by providing odds ratio (OR) with the corresponding 95% CI for each model. The receiver operating characteristic (ROC) curves were used to determine the best cut-off values. The area under the ROC curve (AUC) and the related 95% CI were calculated. Linear forward regression models with the Bayesian information criterion were also performed to evaluate the relationship between independent variables and monophasic IRVF assumed as the dependent variable. Inter-observer variability was tested by intraclass correlation coefficient (ICC) in two-way models for agreement (strength of absolute agreement between raters was considered poor, fair, moderate, strong, or almost perfect according to ICC values < 0.30, 0.3-0.49, 0.50-0.69, 0.70-0.89, and ≥0.90, respectively).
Statistical significance was assumed at p < 0.05 with a two-tailed null hypothesis. The statistical analyses were carried out by using SPSS (version 27.0; SPSS Inc., Chicago, IL, USA) and R environment (version 4.0.3; R Foundation for Statistical Computing, Vienna, Austria).

Results
A total of 200 patients met the inclusion criteria. After exclusion criteria, 173 patients were included in the analysis (Table 1) and divided into two groups, depending on whether they did (n = 55) or did not (n = 118) develop postoperative complications within the first 7 days. The demographics and clinical characteristics of the patients are illustrated in Table 1.

Postoperative AKI
Patients who developed postoperative AKI had more cardiovascular comorbidities treated by oral β-blockers with slightly higher baseline RDRI, higher intra-abdominal pressure due to surgical pneumoperitoneum, lower renal filtration gradient, and lower hemoglobin levels in the postoperative period (Table 2).  On postoperative day 1, the IRVF pattern was continuous in 22 patients (12.7%), biphasic in 80 (46.2), and monophasic in 71 (41%). The median RDRI was 0.69 (IQR 0.63-0.74). As compared with preoperative measurements, the IRVF pattern worsened in 54 patients (31%), changing from continuous to biphasic in 10, from continuous to monophasic in 8, and from biphasic to monophasic in 36, suggestive of a trend toward stages of more severe congestion.

Postoperative AKI
Patients who developed postoperative AKI had more cardiovascular comorbidities treated by oral β-blockers with slightly higher baseline RDRI, higher intra-abdominal pressure due to surgical pneumoperitoneum, lower renal filtration gradient, and lower hemoglobin levels in the postoperative period (Table 2).
In multivariate analysis, IRVF pattern and RDRI were the only factors independently associated with postoperative AKI (Table 3). Kaplan-Meier analysis showed a significantly shorter time to the first complication with monophasic than non-monophasic IRVF patterns ( Figure 3).   In multivariate analysis, IRVF pattern and RDRI were the only factors independently associated with postoperative AKI (Table 3). Kaplan-Meier analysis showed a significantly shorter time to the first complication with monophasic than non-monophasic IRVF patterns ( Figure 3).

Postoperative Complications
Of the 55 patients with various types of complications (Table 4), no one died or required intensive care unit admittance. Patients who developed any complications within day 7, compared with patients who did not (Table 5), had a higher intra-abdominal pressure due to surgical pneumoperitoneum, lower abdominal perfusion pressure, lower renal filtration gradient, and a tendency for greater blood loss requiring more fluid administration in the intraoperative period with lower hemoglobin levels in the postoperative period.   In addition, they had more frequent IRVF worsening (58% vs. 19% p < 0.001) ( Table 5) (Table 5). In multivariate analysis, the IRVF pattern and RDRI were the only factors independently associated with postoperative complications (Table 6). Kaplan-Meier analysis showed a significantly shorter time to the first complication with monophasic than non-monophasic IRVF patterns ( Figure 5).

Postoperative Monophasic IRVF
At univariate analysis, the development of monophasic IRVF was associated with high pneumoperitoneum pressure and a large amount of fluid given during the intraoperative period (Table 7), but not mean arterial pressure. In the multivariate analysis, the intra-abdominal pressure remained the only factor independently associated with the development of monophasic IRVF (Table 8).  operative RDRI (0.74 (0.73-0.77) vs. 0.67 (0.61-0.71) p < 0.001) with the best cut-off of 0.72 (AUROC 0.73, CI: 0.65-0.81, p < 0.001). Inferior vena cava maximum diameter and collapsibility index were not significantly different between the two groups ( Table 5). In multivariate analysis, the IRVF pattern and RDRI were the only factors independently associated with postoperative complications (Table 6). Kaplan-Meier analysis showed a significantly shorter time to the first complication with monophasic than non-monophasic IRVF patterns ( Figure 5).

Arterial-to-Venous Coupling and Development of Complications
From the present cohort, combining arterial and venous Doppler spectra, we can identify six different perfusion profiles: (1) "normal", with preserved arterial perfusion without congestion characterized by an RDRI < 0.72 and continuous IRVF; (2) "pure hypoperfused" with RDRI > 0.72 and continuous IRVF; (3) "compensated congestive profile" with biphasic IRVF but RDRI < 0.72; (4) "mixed mild hypo-perfused and decompensated congestive profile" with biphasic IRVF and RDRI > 0.72; (5) "pure severe congestive profile" with monophasic IRVF but normal RDRI < 0.72; and (6) "mixed severe hypo-perfusedcongestive profile" with monophasic IRVF and increased RDRI > 0.72. (Figure 6). without congestion characterized by an RDRI < 0.72 and continuous IRVF; (2) "pure hypoperfused" with RDRI > 0.72 and continuous IRVF; (3) "compensated congestive profile" with biphasic IRVF but RDRI < 0.72; (4) "mixed mild hypo-perfused and decompensated congestive profile" with biphasic IRVF and RDRI > 0.72; (5) "pure severe congestive profile" with monophasic IRVF but normal RDRI < 0.72; and (6) "mixed severe hypo-perfused-congestive profile" with monophasic IRVF and increased RDRI > 0.72. (Figure 6). The frequency distribution of these profiles was associated with the occurrence of postoperative AKI or any complications (Table 9). The most frequent profile (number 3) was present in 36% of patients and associated with a low rate of complications (5%). Profiles with arterial hypoperfusion but no congestion (numbers 2 and 4) were associated with a 28-33% complication rate. Profiles with severe venous congestion were associated with complication rates from 46% in the case of isolated venous congestion to 76% in the case of concomitant arterial hypoperfusion. Kaplan-Meier estimates of the time to first postoperative AKI or complication for the six flow profiles are represented, respectively, in Figures 7 and 8. The frequency distribution of these profiles was associated with the occurrence of postoperative AKI or any complications (Table 9). The most frequent profile (number 3) was present in 36% of patients and associated with a low rate of complications (5%). Profiles with arterial hypoperfusion but no congestion (numbers 2 and 4) were associated with a 28-33% complication rate. Profiles with severe venous congestion were associated with complication rates from 46% in the case of isolated venous congestion to 76% in the case of concomitant arterial hypoperfusion. Kaplan-Meier estimates of the time to first postoperative AKI or complication for the six flow profiles are represented, respectively, in Figures 7 and 8.

Reliability
The IRVF patterns were identifiable in all cases with complete inter-observer agreement. RDRI could not be determined in one or in both kidneys of 10 subjects (5%) due to poor acoustic windows caused by intestinal gas, presence of surgical drainage or obesity. The inter-rater agreement for RDRI was strong (ICC 0.844 (95% CI from 0.792 to 0.844)).

Reliability
The IRVF patterns were identifiable in all cases with complete inter-observer agreement.
RDRI could not be determined in one or in both kidneys of 10 subjects (5%) due to poor acoustic windows caused by intestinal gas, presence of surgical drainage or obesity. The inter-rater agreement for RDRI was strong (ICC 0.844 (95%CI from 0.792 to 0.844)).

Discussion
In this study, the prediction of AKI and other postoperative complications after major urologic laparoscopic surgery was researched using a renal point-of-care ultrasound. The main findings were that (1) a postoperative monophasic IRVF pattern predicts AKI or any complications by day 7, (2) high intraoperative abdominal pressure represents the only independent determinant of postoperative monophasic IRVF pattern, and (3) the overall number of postoperative AKI or any complication relies on the balance between arterial inflow and venous outflow in order to ensure the adequacy of peripheral perfusion.

Discussion
In this study, the prediction of AKI and other postoperative complications after major urologic laparoscopic surgery was researched using a renal point-of-care ultrasound. The main findings were that (1) a postoperative monophasic IRVF pattern predicts AKI or any complications by day 7, (2) high intraoperative abdominal pressure represents the only independent determinant of postoperative monophasic IRVF pattern, and (3) the overall number of postoperative AKI or any complication relies on the balance between arterial inflow and venous outflow in order to ensure the adequacy of peripheral perfusion.
Postoperative IRVF patterns have not been extensively studied in the perioperative period of laparoscopic surgery. This study shows that IRVF assessment is feasible with a widely available conventional ultrasonographic system and can provide information on the risk of postoperative AKI as well as any complications within the postoperative day 7. Patients who developed postoperative monophasic IRVF patterns were those who received larger amounts of intraoperative fluids and had higher intra-abdominal pressure, independently of mean arterial pressure. Although a low mean arterial pressure may represent a condition affecting intrarenal circulation, the lack of significant association with changes in IRVF patterns in the present study suggests that mechanisms other than the sole blood supply could be involved in the development of renal dysfunction. Some of these might be the increase in venous resistance to flow induced by high intra-abdominal pressure and reduction of parenchymal compliance due to increased renal interstitial pressure caused by fluid overload; both mechanisms ultimately lead to a "splanchnic tamponade" [20]. This is an important issue to be considered because introduces the concept that splanchnic tamponade due to venous congestion is not necessarily caused by fluid overload but may be due to other conditions affecting vascular compliance such as abdominal hypertension or abdominal compartment syndrome, leading to the development of the new concept of "dry" splanchnic tamponade.
Most of the ultrasonographic studies on congestion came from the application of VEXUS protocols in cardiologic patients with varying degrees of right and/or left cardiac dysfunction and concomitant fluid overload [9]. However, our results, collected in a population of elective surgical patients submitted to a strict ERAS protocol without known cardiac pathologies or signs of heart decompensation, showed postoperative Doppler signs of splanchnic congestion even in the absence of dilatation of the inferior vena cava, thus suggesting the absence of fluid overload. These findings support the hypothesis that in our patients the main pathophysiological mechanism causing congestion may be due to a "restrictive venous compliance" rather than right heart dysfunction or overt fluid overload. If our findings would be confirmed in further studies, the VEXUS ultrasound approach as originally described, should be rethought avoiding considering the inferior vena cava as "a gate" in order to decide whether or not to proceed with the assessment of congestion in the splanchnic compartment. This consideration would be extremely important in the context of critically ill patients, due to the multiple pitfalls of IVC ultrasonography in this setting [21].
Furthermore, in our study, a considerable number of patients in the postoperative period presented with venous patterns suggestive of mild-to-moderate congestion, without actually developing postoperative AKI or any complications. These findings support the hypothesis that venous congestion may be effectively counterbalanced by a proper cardiac output and arterial in-flow [22].
Moreover, the monophasic IRVF pattern was associated with worst outcomes compared with the continuous or biphasic patterns; these findings are supported by the influence of splanchnic congestion on postoperative complications [23][24][25][26] even if the amount of intraoperative fluid administered and pneumoperitoneum pressures were relatively low, according to the ERAS protocols of our center [27]. Even if differences in the intrabdominal pressures during pneumoperitoneum were statistically different between patients who eventually developed postoperative AKI or any complications compared with those who didn't, they were not predictive of AKI or any complication and we must say that the absolute value of these pressures was very close, and thus unhelpful in the clinical practice. Conversely, postoperative AKI and any complications can be predicted by IRVF patterns in the early postoperative period, potentially allowing the timely treatment of individual patients.
RDRI was previously demonstrated to be as a useful tool to assess splanchnic regional perfusion in critically ill patients [10,15,16]. The present study confirms and expands these findings, as an impaired postoperative RDRI was independently associated with a strong odds ratio with postoperative AKI and any complications.
Our study is different from all those previously published on the topic as for the first time it simultaneously evaluates arterial perfusion, venous congestion, and their mutual interrelationship. The combination of overt arterial hypoperfusion (RDRI > 0.72) and moderate-to-severe venous congestion (biphasic or monophasic venous patterns) was associated with the worst outcomes, suggesting that only the concomitant assessment of the arterial and venous side of the circulation may provide the entire picture of the adequacy of end-organ perfusion requirements.
Hopefully, this concept will pave the way for designing a large randomized clinical trial to investigate if and how addressing the adequacy of splanchnic perfusion more precisely by US can improve the postoperative outcomes of these patients.
We acknowledge some study limitations. This is a single-center study from a referral center. It has inherent flaws related to selection and referral bias. Our work was not aimed to evaluate the different effects on splanchnic circulation by different types of urological surgery that may be important factors affecting perfusion. Cardiac output, the major determinant of splanchnic perfusion, was not measured because it does not comply with our ERAS protocol in this type of surgery. The results of the present study refer to elective laparoscopic urological surgery and therefore cannot be transposed to other types of surgery.

Conclusions
The concomitant assessment of the arterial (RDRI) and venous (IRVF) side of the circulation predicts postoperative complications after major laparoscopic urologic surgery and may be considered a useful ultrasonographic biomarker to stratify vulnerable patients at risk for the development of postoperative complications. Whether prediction of AKI and other complications can anticipate or ameliorate the treatment of perfusion disorders, improving outcomes requires further investigation.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jcm12155013/s1, Figure S1: Flow diagram of the study. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: Data will be made available by the corresponding author for global collaboration on reasonable request, within the national restrictions imposed by privacy laws and ethics.