Can Activities of Daily Living Predict Complications following Percutaneous Nephrolithotomy?

Can Activities of Daily Living Predict Complications following Percutaneous Nephrolithotomy? David A. Leavitt,* Piruz Motamedinia, Shamus Moran, Michael Siev, Philip T. Zhao, Nithin Theckumparampil, Mathew Fakhoury, Sammy Elsamra, David Hoenig, Arthur Smith and Zeph Okeke From the Smith Institute for Urology, Hofstra North Shore-Long Island Jewish School of Medicine (DAL, PM, SM, MS, PTZ, NT, MF, SE, DH, AS, ZO), New Hyde Park, New York, Vattikuti Urology Institute, Henry Ford Health System (DAL), Detroit, Michigan, Department of Urology, Yale University (PM), New Haven, Connecticut, and Rutgers Cancer Institute (SE), New Brunswick, New Jersey

Purpose: Activities of daily living provide information about the functional status of an individual and can predict postoperative complications after general and oncological surgery. However, they have rarely been applied to urology. We evaluated whether deficits in activities of daily living could predict complications after percutaneous nephrolithotomy and how this compares with the Charlson comorbidity index and the ASAÒ (American Society of AnesthesiologistsÒ) classification. Materials and Methods: We retrospectively reviewed the records of all patients who underwent percutaneous nephrolithotomy between March 2013 and March 2014. Those with complete assessment of activities of daily living were included in analysis. Perioperative outcomes, complications and hospital length of stay were examined according to the degree of deficits in daily living activities. Results: Overall 176 patients underwent a total of 192 percutaneous nephrolithotomies. Deficits in activities of daily living were seen in 16% of patients, including minor in 9% and major in 7%. Complications developed more frequently in those with vs without deficits in daily living activities (53% vs 31%, p ¼ 0.029) and length of stay was longer (2.0 vs 4.5 days, p ¼ 0.005). On multivariate logistic regression activities of daily living were an independent predictor of complications (OR 1.11, p ¼ 0.01) but ASA classification and Charlson comorbidity index were not. Conclusions: Activities of daily living are easily evaluated prior to surgery. They independently predict complications following percutaneous nephrolithotomy better than the Charlson comorbidity index or the ASA classification. Preoperative assessment of daily living activities can help risk stratify patients and may inform treatment decisions.

Abbreviations and Acronyms ADL ¼ activity of daily living BMI ¼ body mass index CCI ¼ Charlson comorbidity index PCNL ¼ percutaneous nephrolithotomy Accepted for publication December 18, 2015. No direct or indirect commercial incentive associated with publishing this article. The corresponding author certifies that, when applicable, a statement(s) has been included in the manuscript documenting institutional review board, ethics committee or ethical review board study approval; principles of Helsinki Declaration were followed in lieu of formal ethics committee approval; institutional animal care and use committee approval; all human subjects provided written informed consent with guarantees of confidentiality; IRB approved protocol number; animal approved project number. * Correspondence: Vattikuti Urology Institute, Henry Ford Hospital, 2799 West Grand Blvd., K9, Detroit, Michigan 48202 (telephone: 313-9162066; FAX: 313-916-7577; e-mail: david.a. [email protected]).

Key Words: kidney calculi; activities of daily living; nephrostomy, percutaneous; postoperative complications; risk THERE is considerable interest across the surgical community in the ability to predict which patients are at highest risk for perioperative complications.1e6 In urology this is particularly true for PCNL, after

which complications are common at 20% to 50% of cases and occasionally life threatening.4,7e9 Indeed, myriad studies have attempted to determine predictors of PCNL related complications.10e12

0022-5347/16/1956-1805/0 THE JOURNAL OF UROLOGY® Ó 2016 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.

http://dx.doi.org/10.1016/j.juro.2015.12.076 Vol. 195, 1805-1809, June 2016 Printed in U.S.A.

www.jurology.com

j

1805

1806

CAN ACTIVITIES OF DAILY LIVING PREDICT COMPLICATIONS AFTER NEPHROLITHOTOMY

CCI, which represents an aggregate of preexisting chronic diseases, and ASA score have been shown to predict postoperative complications in a variety of settings, although use in endourology and specifically for PCNL has been limited. However, these metrics are imperfect and incomplete since ASA score relies partly on assessor subjectivity. In addition, neither ASA score nor CCI objectively incorporates patient functional status, which is becoming recognized as one of the most potent predictors of complications.2,6,13 Assessment of ADLs, which can provide basic and objective information on individual functional status, has been in use for more than a half century.14 Moreover, ADLs do not rely on subjectivity and are easy to evaluate. In use in the geriatric literature for aging, in recent years ADL assessment has been applied to surgical oncology. It has repeatedly proved to be a reliable predictor of complications, hospital stay and discharge destination. However, to our knowledge ADLs have not been widely applied to urology or at all to PCNL. To this end we sought to determine whether deficits in ADLs could predict post-PCNL complications and how this would compare with CCI and ASA.

MATERIALS AND METHODS In this institutional review board approved, retrospective study we reviewed all PCNL procedures performed between March 2013 and March 2014 at a single institution by 3 experienced endourologists (DH, AS and ZO). ADLs were assessed by an internist along with history and physical examination as part of preoperative testing in all patients scheduled for elective surgery at our institution. Assessment was done within 30 days of surgery. We included patients in whom a full ADL assessment was performed preoperatively. Patients without a complete ADL assessment and those younger than 18 years were excluded from analysis. Patient demographics, perioperative outcomes, ASA score, CCI and complications were collected from patient charts. The ADL evaluation assessed patient ability to perform basic functional activities that are considered the essential elements of self-care and have been deemed necessary for independent survival.14 Individual ADL tasks included ambulation, transferring, toileting, bathing, dressing, eating, communicating and swallowing. Communicating assessed patient difficulty with or complete inability to understand and/or speak while swallowing gauged difficulty with consuming liquids and foods. Each individual ADL was scored on a scale between 0dno deficit and complete independence, and 4dfull deficit and complete dependence. In addition, a total ADL score was calculated by summing the score of each individual ADL with 0 representing complete independence in all ADLs and 32 representing complete dependence in all ADLs. Composite deficits in ADLs were stratified into 3 groups, including 1) no deficit in any ADL category (no

deficit), 2) at least a single score of 1 in any ADL category but no score greater than 1 in any category (minor deficit) and 3) at least a single score of 2 in any ADL category (major deficit). CCI was calculated based on patient comorbid conditions present at the same time that ADLs were assessed.15 ASA classification was determined on the day of surgery by the anesthesiologist overseeing the surgery.16,17 All PCNLs were performed with the patient prone. Percutaneous renal access was achieved by the urologist except in patients with a preexisting nephrostomy catheter. Access tracts were dilated to 30Fr with NephroMaxÔ balloon dilators or Amplatz renal dilators (CookÒ Medical) according to surgeon preference. A CyberWandÒ ultrasonic lithotripter and a 24Fr rigid nephroscope were used. Postoperative renal drainage catheter type was determined according to surgeon judgment. Operative time included cystoscopy, ureteral catheterization, prone repositioning, renal access, stone removal and drain placement. Stone size was calculated as the largest diameter in any dimension of the single largest stone. All complications within 30 days of surgery were included and graded according to the modified Clavien-Dindo classification as minordClavien I or II, or majordClavien IIIa or greater.18 For patients treated with staged procedures during the same hospitalization length of stay was calculated from the date of the last procedure performed. Patients with contaminated preoperative urine cultures were considered to have positive urine cultures because many of them received preoperative antibiotics as if the culture were positive. All statistical analyses were performed using STATAÒ/ IC, version 11.0 with p <0.05 considered significant. Continuous variables were compared across the 3 ADL substratifications using ANOVA with the Bonferroni post hoc test. Categorical variables were compared using the Fisher exact test when appropriate. Univariate and multivariate logistic regression analyses were performed to determine predictors of a perioperative event and ORs.

RESULTS We identified 176 patients who underwent a total of 192 PCNLs, including 16 staged procedures. The table shows patient characteristics. A majority of patients with major ADL deficits were women. Of the entire study cohort 7% showed major deficits, 9% had minor deficits and 84% showed no deficits. Mean BMI was 29.9 kg/m2 in the entire study group. Mean stone size, and the proportion of partial and full staghorn stones did not significantly differ among the groups. Overall mean ASA score was 2.4 with a significantly lower average ASA score in patients with no ADL deficits compared to those with minor and major deficits (2.3 vs 2.8 and 3.1, respectively, p <0.001). A similar result was seen for CCI, including for no ADL deficits, a minor deficit and a major deficit (0.71, 1.5 and 1.8, respectively, p <0.001). Complications developed in 61 of 176 patients (35%), including 31% with no ADL deficits, 47%

CAN ACTIVITIES OF DAILY LIVING PREDICT COMPLICATIONS AFTER NEPHROLITHOTOMY

1807

Patient characteristics ADL Deficit

No. pts (%) Mean age (range) No. male (%) Mean kg/m2 BMI (range) Mean stone size (mm) No. staghorn (%): Partial Full No. pos preop urine culture (%) Mean ASA score (range) Mean CCI (range) Mean operative time (mins) No. staged (%)

Overall

None

Minor

Major

176 57.7 (24e97) 96 (55) 29.9 (16.6e49.3) 21.1

148 (84) 56.4 (24e90) 84 (57) 29.8 (16.6e49.3) 21.8

15 (9) 67.7 (49e84)* 8 (53) 33.3 (19.8e47.9) 17.9

13 (7) 61.4 (34e97) 4 (31) 26.7 (19.8e35.1)† 20.3

34 5 77 2.4 0.86 103 15

31 4 62 2.3 0.71 104 11

2 0 9 2.8 1.47 92 2

(19) (3) (44) (1e4) (0e6) (9)

(21) (3) (42) (1e4) (0e6) (7)

(13) (60) (2e3)* (0e5)* (13)

1 1 6 3.1 1.8 109 2

p Value

(8) (8) (46) (3e4)* (0e4)* (15)

e 0.006 0.19 0.02 0.35 0.49 0.41 <0.001 <0.001 0.53 0.25

* Significantly different vs no deficit (ANOVA followed by Bonferroni post hoc test). † Significantly different vs minor deficit (ANOVA followed by Bonferroni post hoc test).

with minor ADL deficits and 62% with major ADL deficits (p ¼ 0.05). Complications were significantly more likely in patients with any deficit in ADL compared to those with no deficit (53% vs 31%, p ¼ 0.029). Univariate analysis of individual ADL categories showed that ambulation, toileting, bathing and dressing were predictive of complications. However, all lost individual significance on multivariate analysis. No significant differences in major complications (Clavien IIIa or greater) or transfusion rates were found. Intensive care unit level care occurred more frequently in patients with minor and/or major ADL deficits compared to those without deficits (13% and 8%, respectively, vs 2%, p ¼ 0.05). Sepsis developed more often as ADL deficits increased, including in 2% of patients with no ADL deficit, in 13% with a minor ADL deficit and in 23% with a major ADL deficit (p ¼ 0.003). On univariate analysis total ADL score predicted overall complications as a continuous variable (OR 1.11, p ¼ 0.01) and as a categorical variable (OR 2.01, p ¼ 0.01). ASA score (OR 1.75, p ¼ 0.033) and positive preoperative urine culture (OR 1.91, p ¼ 0.045) also predicted overall complications while CCI did not (OR 1.04, p ¼ 0.76). Age, staged procedures, gender and staghorn stone configuration did not predict complications. Deficits in ADLs were strongly predictive of sepsis (OR 1.15, p <0.001) while ASA score, CCI and positive preoperative urine culture were not. Neither ADLs, ASA score nor CCI was predictive of transfusion or intensive care unit admission. On multivariate logistic regression including ASA score, CCI, ADL sum and preoperative urine culture status ADL sum independently predicted complications (OR 1.10, p ¼ 0.02). ASA score, CCI and preoperative urine culture status did not. Duration of hospitalization correlated with the degree of deficit in ADLs. Mean hospital stay was

significantly longer in patients with any ADL deficit compared to those with no deficit (4.5 vs 2.9 days, p ¼ 0.005). Similarly those with major deficits in ADLs experienced longer hospitalization than those with no or minor ADL deficits (5.4 vs 3.0 days, p ¼ 0.002). Median length of stay was longer in patients with any deficit compared to those with no deficit (3 vs 2 days). Hospitalization greater than 1 week occurred more frequently as ADL deficits increased, including in 7% of patients with no deficit, 13% with a minor deficit and 31% with a major deficit.

DISCUSSION Preoperative functional status as assessed by ADLs can be used to predict complications following PCNL. Patients with greater ADL deficits were more likely to experience complications, sepsis and longer hospitalization. Although individual ADL tasks were related to the risk of complications on univariate analysis, none proved more important than another on multivariate analysis. This may reflect a degree of collinearity among individual ADL tasks. Approximately 1 of 7 patients in the entire study group had at least a minor deficit in ADL, suggesting that suboptimal functional status may be common in patients undergoing PCNL. Staghorn stone configuration and stone size did not significantly differ among the groups, suggesting that differences in complication rates cannot be fully explained by stone characteristics. Moreover, age alone did not explain these findings as the average age in those with minor deficits was actually higher than in those with major deficits. This is in line with other studies showing that age alone is not an independent predictor of complications.19 BMI in patients with minor ADL deficits was significantly higher than in those with major deficits, which may reflect the higher degree of frailty and accompanying body mass loss in those with major deficits.

1808

CAN ACTIVITIES OF DAILY LIVING PREDICT COMPLICATIONS AFTER NEPHROLITHOTOMY

The overall frequency of complications was 35%, of which 80% were minor in accordance with other reports.4,5,8,9 This emphasizes that complications following PCNL remain common. Complications developed in 61% of patients with major ADL deficits, 47% with minor deficits and 31% without deficits (p ¼ 0.05). A trend toward more frequent major complications was seen in those with any ADL deficit compared to those without a deficit (p ¼ 0.09). Length of hospital stay significantly increased as ADL deficits increased, mirroring findings in another prospective study of patients with cancer.20 Interestingly when reviewing the records of all patients who experienced a complication, no statistically significant difference in length of stay was found according to the degree of ADL deficit. Nonetheless, a trend toward longer hospital stay was seen in those with major ADL deficits relative to those with no or minor ADL deficits (7.3 vs 4.5 days, p ¼ 0.06). Indeed, others have noted that deficits in functional status can predict discharge to facilities other than the patient home.21 Counseling patients and family members accordingly about postoperative expectations may provide an opportunity to better forecast length of stay and discharge needs, and minimize unnecessary inpatient hospital days. In this study CCI was not predictive of complications following PCNL, which contrasts with other published findings.4,5,22 A plausible explanation becomes more apparent after thoroughly reviewing these studies. Moreno-Palacios et al found that a high CCI (5 or greater) independently predicted overall and high grade complications (Clavien III or greater) but no relationship existed in patients with moderate or low CCI scores.5 Importantly only approximately 3% of that study population were in the high CCI group, questioning its generalizability. The other 2 studies, which apparently used the same data set, showed that a high CCI (2 or greater) was significantly associated with medical but not with overall or surgical complications following PCNL.4,22 Similar results were seen in a recent review of major surgery in patients with cancer, in which CCI did not predict complications.13 All in all it appears that functional status such as ADLs may be more useful than CCI to predict complications. Since it was first introduced in 1941, the ASA score has repeatedly proven to be a useful predictor of perioperative complications and mortality.13,16,17 A recent review focusing on urological surgeries using the NSQIP (National Surgical Quality Improvement Program) database corroborated that a higher ASA class and dependent functional status were independently associated with postoperative complications.1 In the current study ASA score

predicted complications on univariate analysis but not on multivariate analysis while ADL remained significant on univariate and multivariate analyses. A possible cause may be that most patients in this study were in the ASA 2 or 3 classification with only 13 considered ASA class 1 and only 2 considered ASA class 4. Another frequently cited concern with the ASA classification system is its inherent subjectivity. In fact, in studies that attempted to answer this question the interrater reliability appeared to be modest at best (k index approximately 0.2 to 0.4) while ADLs appeared to fare better.23e25 Thus, in a sample of patients such as those in the current series, of whom most were in the ASA 2 or 3 category, ADLs may prove even more discriminatory than ASA class to winnow out those at highest risk for complications. A robust body of evidence is mounting across myriad surgical and medical fields that illustrates the influence that deficits in ADL can have on a multitude of factors. ADLs have predicted higher surgical complications, increased 1-year mortality in elderly patients following hospitalization, higher hospital costs, longer length of hospital stay, more frequent surgical site infections, and less successful ventilator weaning and survival in patients with chronic obstructive pulmonary disease.3,26e29 The usefulness of preoperative ADL assessment or another simple evaluation of functional status lies in its ability to quickly risk stratify PCNL candidates and possibly influence treatment decisions. Patients with major ADL deficits may be better served by a treatment other than PCNL as such individuals may not have enough physiological reserve to endure and successfully recover from a major operation. Limitations of this study include its retrospective, single center design. It is possible that some complications were missed despite exhaustive efforts to capture them. Further, the findings in this series may have limited generalizability as all patients were treated by endourologists at a tertiary referral, high volume stone center. Another limitation is our inability to completely tease out the history of preoperative infectious and uroseptic events in some referred patients. Postoperative complications and outcomes were evaluated out to 30 days while others have evaluated 90-day outcomes. Nonetheless, we believe that assessing patient functional status is an important factor for estimating perioperative risk and gauging expected recovery. In the current medical environment physician and health system compensation is becoming increasingly tied to patient outcomes. We hope that appropriate expectations could be prospectively established using a system to objectively

CAN ACTIVITIES OF DAILY LIVING PREDICT COMPLICATIONS AFTER NEPHROLITHOTOMY

decipher patient complexity and how it may affect outcomes.

CONCLUSIONS ADLs are simple to assess preoperatively. Deficits in ADLs are not uncommon in patients undergoing PCNL and these deficits independently predict postoperative complications better than CCI or ASA score. Moreover, patients with significant

1809

deficits in ADLs can expect significantly longer hospitalization and a higher need for specialized postoperative care. Preoperative assessment of ADLs appears to be a useful tool to risk stratify candidates for PCNL. This information may help inform treatment decisions by better identifying patients at highest risk for complications in whom preoperative intervention to curtail surgical risks may be most useful.

REFERENCES 1. Patel HD, Ball MW, Cohen JE et al: Morbidity of urologic surgical procedures: an analysis of rates, risk factors, and outcomes. Urology 2015; 85: 552. 2. Makary MA, Segev DL, Pronovost PJ et al: Frailty as a predictor of surgical outcomes in older patients. J Am Coll Surg 2010; 210: 901. 3. Fukuse T, Satoda N, Hijiya K et al: Importance of a comprehensive geriatric assessment in prediction of complications following thoracic surgery in elderly patients. Chest 2005; 127: 886. 4. Unsal A, Resorlu B, Atmaca AF et al: Prediction of morbidity and mortality after percutaneous nephrolithotomy by using the Charlson Comorbidity Index. Urology 2012; 79: 55.

11. de la Rosette JJ, Zuazu JR, Tsakiris P et al: Prognostic factors and percutaneous nephrolithotomy morbidity: a multivariate analysis of a contemporary series using the Clavien classification. J Urol 2008; 180: 2489. 12. Wang Y, Jiang F, Wang Y et al: Post-percutaneous nephrolithotomy septic shock and severe hemorrhage: a study of risk factors. Urol Int 2012; 88: 307. 13. Feng MA, McMillan DT, Crowell K et al: Geriatric assessment in surgical oncology: a systematic review. J Surg Res 2015; 193: 265. 14. Staff of the Benjamin Rose Hospital: Multidisciplinary studies of illness in aged persons. II. A new classification of functional status in activities of daily living. J Chronic Dis 1959; 9: 55.

5. Moreno-Palacios J, Maldonado-Alcaraz E, Montoya-Martinez G et al: Prognostic factors of morbidity in patients undergoing percutaneous nephrolithotomy. J Endourol 2014; 28: 1078.

15. Charlson ME, Pompei P, Ales KL et al: A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987; 40: 373.

6. Kim SW, Han HS, Jung HW et al: Multidimensional frailty score for the prediction of postoperative mortality risk. JAMA Surg 2014; 149: 633.

16. Saklad M: Grading of patients for surgical procedures. Anesthesiology 1941; 2: 281.

7. Michel MS, Trojan L and Rassweiler JJ: Complications in percutaneous nephrolithotomy. Eur Urol 2007; 51: 899. 8. de la Rosette J, Assimos D, Desai M et al: The Clinical Research Office of the Endourological Society Percutaneous Nephrolithotomy Global Study: indications, complications, and outcomes in 5803 patients. J Endourol 2011; 25: 11. 9. Kyriazis I, Panagopoulos V, Kallidonis P et al: Complications in percutaneous nephrolithotomy. World J Urol 2015; 33: 1069. 10. Palmero JL, Nuno de la Rosa I, Miralles J et al: Study of predictive factors for complications after percutaneous nephrolithotomy according to the Clavien classification. Actas Urol Esp 2013; 37: 412.

17. Dripps RD: New classification of physical status. Anesthesiology 1963; 24: 111. 18. Dindo D, Demartines N and Clavien PA: Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004; 240: 205. 19. Okeke Z, Smith AD, Labate G et al: Prospective comparison of outcomes of percutaneous nephrolithotomy in elderly patients versus younger patients. J Endourol 2012; 26: 996. 20. Audisio RA, Pope D, Ramesh HS et al: Shall we operate? Preoperative assessment in elderly cancer patients (PACE) can help. A SIOG surgical task force prospective study. PACE Participants. Crit Rev Oncol Hematol 2008; 65: 156. 21. Badgwell B, Stanley J, Chang GJ et al: Comprehensive geriatric assessment of risk

factors associated with adverse outcomes and resource utilization in cancer patients undergoing abdominal surgery. J Surg Oncol 2013; 108: 182. 22. Resorlu B, Diri A, Atmaca AF et al: Can we avoid percutaneous nephrolithotomy in high-risk elderly patients using the Charlson comorbidity index. Urology 2012; 79: 1042. 23. Sankar A, Johnson SR, Beattie WS et al: Reliability of the American Society of Anesthesiologists physical status scale in clinical practice. Br J Anaesth 2014; 113: 424. 24. Mak PH, Campbell RC, Irwin MG et al: The ASA Physical Status Classification: inter-observer consistency. American Society of Anesthesiologists. Anaesth Intensive Care 2002; 30: 633. 25. Sainsbury A, Seebass G, Bansal A et al: Reliability of the Barthel Index when used with older people. Age Ageing 2005; 34: 228. 26. Chen TY, Anderson DJ, Chopra T et al: Poor functional status is an independent predictor of surgical site infections due to methicillinresistant Staphylococcus aureus in older adults. J Am Geriatr Soc 2010; 58: 527. 27. Walter LC, Brand RJ, Counsell SR et al: Development and validation of a prognostic index for 1-year mortality in older adults after hospitalization. JAMA 2001; 285: 2987. 28. Chuang KH, Covinsky KE, Sands LP et al: Diagnosis-related group-adjusted hospital costs are higher in older medical patients with lower functional status. J Am Geriatr Soc 2003; 51: 1729. 29. Langlet K, Van Der Linden T, Launois C et al: Predictive value of daily living score in acute respiratory failure of COPD patients requiring invasive mechanical ventilation pilot study. BMC Pulm Med 2012; 12: 66.