Mechanical bowel preparation (MBP) used to be the standard of care before major abdominal surgeries with the intent of decreasing bacterial load and thus infection risk [1]. This rationale was thought to be particularly important for any surgery involving bowel resection and anastomoses [1]. MBP was also favored so that the colon would be empty and easier to manipulate. Traditional protocols instruct patients to take oral osmotic agents like polyethylene glycol and sodium phosphate to achieve bowel cleansing [2]. In the context of kidney surgery, although most open surgery (OS) historically involved a retroperitoneal technique, MBP was nonetheless employed due to convention and a theoretical decrease in infection risk.

First within the field of colorectal surgery, the concept of Enhanced Recovery After Surgery (ERAS) protocol challenged many perioperative practices like MBP [3]. This new paradigm asserted that most cases of perioperative morbidity are linked to metabolic endocrine changes and the stress response to surgery [3]. Newer guidelines tend to omit practices that are perceived as causing unnecessary metabolic stress, including MBP and prolonged preoperative fasting.

The shift toward laparoscopic and robotic approaches in the minimally invasive surgery (MIS) era occurred concurrently with the introduction of ERAS protocols, and the transperitoneal approach became commonly employed [3,4]. This transition to the MIS era has not only influenced incision size and location but has also reshaped the entire surgical technique, anatomical approach, and notably, the physiological experience of patients’ abdominal organs, including factors like bowel mobilization, pneumoperitoneum, and surgery duration. Recent guidelines reflect these changes and typically exclude practices perceived to induce unnecessary metabolic stress, and thus, prior to nephrectomy, MBP is no longer recommended under ERAS protocols [5–7]. Using a contemporary national surgical database, the herein-reported study assessed the use of MBP prior to nephrectomy to determine whether there is evidence for its continued use in the MIS era.

Study data and sample

The National Surgical Quality Improvement Program (NSQIP) database was queried for this study. NSQIP is a nationwide collaborative database designed to measure 30-day surgical outcomes for various surgical procedures in the United States. Trained reviewers at over 600 sites record preoperative data through 30-day outcomes from randomly assigned patients. NSQIP data is publicly available and patient information is de-identified. All partial and radical nephrectomies reported in the NSQIP database between 2019 and 2021 were selected for analysis. Inclusion criteria included cases completed by one of three surgical approaches: planned OS, planned laparoscopic surgery, and planned robotic surgery.

Variables

Preoperative and demographic variables included age, sex, body mass index (BMI), diabetes mellitus, smoking status, American Society of Anesthesiologists (ASA) classification, and MBP. Outcomes included postoperative ileus, 30-day reoperation, 30-day readmission, any surgical site infection, operative time, pneumonia, deep venous thrombosis (DVT), pulmonary embolism (PE), and acute renal failure. Postoperative ileus was defined by NSQIP as inability of the patient to sustain themselves with oral intake more than 3 days postoperatively or requiring a nasogastric tube on postoperative day 4 or later. Surgical site infection was defined as presence of either a superficial, deep incisional, or organ space infection within 30 days postoperatively. Renal insufficiency was defined as a rise in serum creatinine >2 mg/dL relative to preoperative values or a new dialysis requirement postoperatively.

Data processing and statistical analyses

Cases with missing information on relevant demographic information or complications were excluded from analysis. Cases were categorized by surgical approach: planned OS vs. planned MIS (including laparoscopic and robotic procedures). Hybrid cases and cases with unplanned conversion to a different approach were also excluded from analyses. Propensity scores were calculated in the Python programming language (version 3.10.12) with the scikit-learn library and included the following covariates: operative time, age, sex, BMI, diabetes mellitus, smoking, and ASA classification I–II (low) vs. III–V (high). Matching was completed by the nearest-neighbor method with 1:3 replacement (k = 3) within the MIS (robotic or laparoscopic approach) group. Each MBP case was matched with three control cases to minimize variance and achieve a sample size approximating the original dataset. Significant differences between groups were identified through chi-square and independent samples t tests as appropriate.

Included cohort and surgical approach

Between 2019 and 2021, a total of 11,869 out of 14,684 (80.8%) nephrectomies met inclusion criteria and were included in the analysis. The majority (69.1%, n = 8,204) utilized MIS. Within the MIS group, 5,357 (65.3%) were completed by a robotic approach and 2,847 (34.7%) were completed by a traditional laparoscopic approach. Open nephrectomy was performed in 3,655 (30.9%) (Fig. 1).

Figure 1. Distribution of surgical approaches from the 2019–2021 National Surgical Quality Improvement Program nephrectomy database with corresponding samples of patients in each group undergoing mechanical bowel preparation (MBP).

Patient characteristics and mechanical bowel preparation

Analysis revealed no significant differences in mean ages (61.2 ± 13.0 years vs. 61.2±12.7 years, p = 0.84) and male sex distribution (60.1% vs. 61.4%, p = 0.17) between the MIS and OS groups. However, the rate of MBP was higher in the MIS group (16.0% vs. 10.0%, p < 0.001). Additionally, patients in the MIS group exhibited lower rates of diabetes mellitus (20.4% vs. 23.2%, p < 0.001) and better ASA classification (low in 40.1% vs. 24.9%, high in 58.9% vs. 75.2%; p < 0.001) (Table 1).

Table 1 . Demographics and preoperative characteristics.

Characteristic	Overalla)	Planned MIS	Planned OS	p valueb)
No. of subjects	11,869	8,204	3,665
Age (yr)	61.2 ± 12.9	61.2 ± 13.0	61.2 ± 12.7	0.84
Male sex	7,179 (60.5)	4,928 (60.1)	2,251 (61.4)	0.17
Obesityc)	5,566 (46.9)	3,820 (46.6)	1,746 (47.6)	0.29
Diabetes mellitus	2,528 (21.3)	1,677 (20.4)	851 (23.2)	<0.001
Smoking	2,092 (17.6)	1,436 (17.5)	656 (17.9)	0.62
ASA classification
Low
I	162 (1.4)	142 (1.73)	20 (0.6)	<0.001
II	4,122 (34.7)	3,233 (39.4)	889 (24.3)
High
III	6,890 (58.1)	4,545 (55.4)	2,345 (64.0)	<0.001
IV	695 (5.9)	284 (3.5)	411 (11.2)
V	0 (0)	0 (0)	0 (0)
Mechanical bowel preparation	1,681 (14.2)	1,315 (16.0)	366 (10.0)	<0.001

Data are expressed as number only, mean ± standard deviation, or number (%)..

MIS, minimally invasive surgery; OS, open surgery; ASA, American Society of Anesthesiologists..

^a)Overall data represents planned minimally invasive (laparoscopic or robotic) and planned open cases combined..

^b)Two-sample t test and chi-square tests were used to determine differences between the MIS and OS groups as appropriate..

^c)Body mass index ≥30 kg/m²..

Associations with 30-day complications

Overall, there were no significant associations between MBP and 30-day complications including postoperative ileus, reoperation, readmission, any surgical site infection, operative time, pneumonia, DVT, PE, or acute renal failure. Stratification by surgical approach (as shown in Table 2) revealed that within the MIS group, the occurrence of MBP was linked to a reduced rate of postoperative ileus, with percentages of 0.9% vs. 1.9%, representing an odds ratio (OR) of 0.48 (95% confidence interval [CI], 0.27–0.87; p = 0.02). Conversely, in the OS group, patients with MBP experienced higher rates of readmission (10.9% vs. 7.3%; OR, 1.5 [95% CI, 1.1–2.2]; p = 0.02) and perioperative infection complications (7.75% vs. 4.1%; OR, 1.9 [95% CI, 1.27–2.96]; p = 0.003) compared to those without MBP.

Table 2 . Sub-analysis of planned MIS approach (robotic or laparoscopic only) vs. planned OS approach.

Variable	Planned MIS					Planned OS
	Overall	MBP	No MBP	p valuea)		Overall	MBP	No MBP	p valuea)
No. of subjects	8,204 (100)	1,315 (16.0)	6,889 (84)	-		3,665 (100)	366 (10.0)	3,299 (90.0)	-
Postoperative ileus	141 (1.7)	12 (0.9)	129 (1.9)	0.02		228 (6.2)	30 (8.2)	198 (6.0)	0.12
30-Day reoperation	138 (1.7)	20 (1.5)	118 (1.7)	0.70		118 (3.2)	6 (1.6)	112 (3.4)	0.10
30-Day readmission	429 (5.2)	58 (4.4)	371 (5.4)	0.17		282 (7.7)	40 (10.9)	242 (7.3)	0.02
Any surgical site infectionb)	191 (2.3)	33 (2.5)	158 (2.3)	0.71		163 (4.4)	28 (7.7)	135 (4.1)	0.003
Operative time (min)	177.8 ± 73.8	168.9 ± 69.0	179.5 ± 74.5	<0.001		181.9 ± 92.6	203.5 ± 98.6	179.5 ± 92.0	<0.001
Pneumonia	78 (1.0)	17 (1.3)	61 (0.9)	0.22		84 (2.3)	8 (2.2)	76 (2.3)	>0.99
DVT	27 (0.3)	6 (0.5)	21 (0.3)	0.54		48 (1.3)	6 (1.6)	42 (1.3)	0.73
Pulmonary embolism	30 (0.4)	8 (0.6)	22 (0.3)	0.18		34 (0.9)	6 (1.6)	28 (0.8)	0.23
Renal insufficiency	58 (0.7)	10 (0.8)	48 (0.7)	0.94		115 (3.1)	6 (1.6)	109 (3.3)	0.12

Data are expressed as number (%) or mean ± standard deviation..

MIS, minimally invasive surgery; OS, open surgery; MBP, mechanical bowel preparation; DVT, deep venous thromboembolism..

^a)Significant differences determined with chi-square and t tests as appropriate to compare MBP vs. no MBP within each approach group..

^b)Superficial, deep incisional, organ space..

Propensity score matching

Propensity score matching by likelihood of receiving MBP within the MIS group yielded two samples with no significant differences in preoperative characteristics or operative time. The matched dataset showed no association between MBP and postoperative ileus, but the lower rate of 30-day readmission in the MBP group became statistically significant (4.4% vs. 6.4%; OR, 0.67 [95% CI, 0.5–0.91]; p = 0.01) (Table 3). After matching, patients in the MBP group also demonstrated higher rates of pneumonia (1.29% vs. 0.46%; OR, 2.86 [95% CI, 1.47–5.56]; p = 0.002) and PE (0.6% vs. 0%, p < 0.001).

Table 3 . Post propensity score matching perioperative outcomes within the MIS group (robotic or laparoscopic only).

Variable	Overall	MBP	No MBP	p valuea)
No. of subjects	5,260	1,315	3,945	NA
Postoperative ileus	66 (1.25)	12 (0.9)	54 (1.4)	0.25
30-Day reoperation	92 (1.8)	20 (1.5)	72 (1.8)	0.54
30-Day readmission	310 (5.9)	58 (4.4)	252 (6.4)	0.01
Any surgical site infectionb)	150 (2.9)	33 (2.5)	117 (.03)	0.44
Operative timec)	-	-	-	-
Pneumonia	35 (0.7)	17 (1.3)	18 (0.5)	0.002
DVT	18 (0.3)	6 (0.5)	12 (0.3)	0.59
Pulmonary embolism	8 (0.2)	8 (0.6)	0 (0)	<0.001
Renal Insufficiency	22 (0.4)	10 (0.8)	12 (0.3)	0.05

Data are expressed as number only or number (%)..

MIS, minimally invasive surgery; MBP, mechanical bowel preparation; NA, not applicable; DVT, deep venous thromboembolism..

^a)Significant differences determined with chi-square and t tests as appropriate to compare MBP vs. no MBP. ^b)Superficial, deep incisional, organ space. ^c)Operative time included as a covariate in propensity score matching..

This study aimed to characterize the use of MBP prior to nephrectomy in the minimally invasive era and determine whether this practice has any association with perioperative outcomes. The contemporary NSQIP database demonstrates that ERAS guidelines, as pertaining to the use of MBP, are followed in 84.0% (MIS) to 90% (OS) of patients. In addition, after controlling demographic factors and operative time for MIS cases, MBP was not associated with complications including postoperative ileus, reoperation, or surgical site infection. Interestingly, these patients with MBP were more likely to develop pneumonia postoperatively and also had slightly lower rates of 30-day readmission. Overall, our findings confirm that MBP has no benefit in OS nephrectomy, and there is likely little benefit in minimally invasive approaches. Within urologic oncology, MBP has been more extensively investigated for patients undergoing radical cystectomy [8,9]. Xu et al. [9] conducted a randomized trial with 86 patients and found no differences in complications with or without MBP. These studies alongside corroborating meta-analysis data from the field of colorectal surgery have led to MBP falling out of favor prior to most urologic surgery [10,11]. Our findings confirm that this omission is likely appropriate even prior to minimally invasive nephrectomy. In addition, we found adverse outcomes with MBP in the OS group that were not present in the minimally invasive group. Further studies are reasonable to investigate how factors specific to MIS (e.g., pneumoperitoneum, more transperitoneal approaches) may mediate these observed differences.

For MIS, our propensity score-matched analysis controlled for operative time and preoperative characteristics and showed a lower rate of 30-day readmission with MBP but no other improvements in measured outcomes. To our knowledge, only one prior study has specifically investigated MBP in nephrectomy [12]. Sugihara et al. [12] also performed a propensity score-matched analysis in a Japanese series of patients undergoing laparoscopic nephrectomy (not distinguishing between partial and radical) and found no effect of MBP on operative time, postoperative complications, or length of hospital stay. The herein reported work differs from their work given our contemporary data with a large sample of robotic nephrectomy cases. Although similar methodologically, our analysis adds confounding intraoperative factors by including operative time as a covariate for our propensity score matching. Controlling for operative time is particularly important to mitigate the effect of confounding due to unknown variables like surgery type, complexity of the case, surgeon experience, or patient factors [13]. However, in accordance with the study by Sugihara et al. [12], our data also suggest little benefit of MBP in minimally invasive nephrectomy. Despite the lower readmission rate in the MBP group, the absolute difference was ~2%, resulting in a number needed to treat 50. This suggests that MBP has a limited clinical role. Before propensity score matching, MBP was associated with fewer events of postoperative ileus. This observation may be related to a poorly understood interaction between MBP and the bowel manipulation required in the minimally invasive transperitoneal approach, but this difference failed to reach statistical significance after propensity score matching.

Furthermore, in the context of MIS, MBP was associated with higher rates of pneumonia and PE after propensity score matching. This finding may be due to MBP serving as a proxy for patients whose perioperative care did not include ERAS protocols, which is a challenging hypothesis to verify. Alternatively, within our study cohort, MBP might indicate a more intricate surgical procedure. ERAS protocols often exclude MBP while also encouraging practices that may be protective against pneumonia or PE, such as early mobilization after surgery [1]. Retrospective reports, including from our own institution, find that patients who are treated according to the ERAS protocols prior to laparoscopic nephrectomy have a reduced length of stay and fewer postoperative complications [5,14]. In a retrospective study of patients undergoing lung resection, an ERAS protocol that included standardized incentive spirometer teaching and early mobilization was associated with a significantly lower risk of postoperative pneumonia [15]. Thus, the increased risk of pneumonia and PE may not be due to MBP in MIS nephrectomy but rather the lack of ERAS protocols.

Our analysis of the 2019–2021 NSQIP database leads us to conclude that there is no benefit of MBP for reducing perioperative complications prior to OS nephrectomy. The complications observed in the MBP group may be associated with more complex cases, characterized by longer operative times and higher rates of readmission, as well as paradoxically higher infection rates. MBP has been abandoned as it has not been proven to yield superior outcomes in significant urological abdominal procedures, including radical cystectomy, which is considered a contaminated procedure due to exposure of bowel contents to the surgical field [16]. Moreover, our findings are consistent with existing literature from other subspecialties, such as gynecology and general surgery, where MBP has been gradually phased out [17]. Furthermore, it is essential to consider that our results are in line with the ERAS paradigm, which recommends against the routine use of MBP prior to nephrectomy.

The primary strength of our study lies in its utilization of a national database designed to maintain high-quality care for surgical patients. This database is a cross-sectional compilation from various institutions reporting select cases, with statistical methodologies specifically tailored to mitigate inherent selection bias in retrospective studies. However, our study also encompasses several inherent limitations. First, NSQIP is a retrospective cross-sectional database compiled by various institutions, which reports select cases. While propensity score matching was used to address potential confounding of known variables, there may still be significant effects from additional unknown variables on the measured outcomes. Second, the NSQIP nephrectomy data encompassed partial nephrectomies, radical nephrectomies, and nephroureterectomies, each with distinct surgical considerations. For instance, nephroureterectomy may entail longer surgery time and is a more complex surgery, and thus these patients may have higher rates of complications. We attempted to adjust for some of these factors by controlling for operative time with propensity score matching, which yielded two groups with similar operative times for comparison. We acknowledge that similar operative times do not eliminate inherent heterogeneity among these procedures however, and the typical patient populations of the various procedures remain a potential confounding factor. Lastly, the NSQIP database may not capture all relevant variables that could impact outcomes, such as variations in surgical techniques (e.g., transperitoneal vs. retroperitoneal approach), perioperative care protocols, and patient comorbidities. Therefore, there may be residual confounding from unmeasured or unaccounted factors that could influence the observed associations. The interaction between MBP and transperitoneal vs. retroperitoneal approach may be particularly important to consider in future studies because there is typically no bowel manipulation in either the open approach or retroperitoneal MIS. There is anticipated heterogeneity in the implementation of ERAS protocols, as well as variability in adherence to its various components across different centers, surgeons, and patients. While this limitation is acknowledged, it still warrants consideration.

In conclusion, the utilization of MBP prior to nephrectomy remains infrequent, evident in both OS and MIS. MBP has demonstrated no clear clinically significant advantage in terms of perioperative morbidity. This contemporary study supports that MBP is likely just inconvenient and uncomfortable, without a parallel benefit, in patients undergoing nephrectomy in the MIS era.

Ethics statement

The Lifespan Institutional Review Board reviewed this study and deemed it research not involving human subjects (Project Package 2208463-1 and Response Package 2208463-2).

Authors’ contributions

Conceptualization: SS, KM, BG, SK, EH, DG

Formal Analysis: SS, KM

Software: AH

Data Curation: AH, CT, RO

Supervision: SK, EH, DG

Writing–original draft: SS, KM, AH

Writing–review & editing: BG, CT, RO, SK, EH, DG

All authors read and approved the final manuscript.

Conflict of interest

All authors have no conflicts of interest to declare.

Funding/support

None.

Data availability

This data is available by request to the National Surgical Quality Improvement Project (https://www.facs.org/quality-programs/data-and-registries/acs-nsqip/participant-use-data-file/).