does walking speed predict postoperative morbidity?

Advances in Surgery j (2014) j–j

ADVANCES IN SURGERY

Does Walking Speed PredictPostoperative Morbidity?

Nicole T. Townsend, MDa, Thomas N. Robinson, MD, MSb,*aDepartment of Surgery, University of Colorado Anschutz Medical Campus, 12631 East 17thAvenue, MS C305, Aurora, CO 80045, USA; bDepartment of Surgery, University of ColoradoAnschutz Medical Campus, 12631 East 17th Avenue, MS C313, Aurora, CO 80045, USA

Keywords

� Geriatric surgery � Frailty � Timed up-and-go � Gait speed� Preoperative risk assessment

Key points

� Slow walking speed correlates with frailty in older adults.� Slow preoperative walking speed predicts increased postoperativecomplications.

� Slow walking speed is easily measured by the timed up-and-go or 5-m/15-feettimed walk.

� Slow walking speed is associated with higher 1-year postoperative mortality.

INTRODUCTIONThe demographic inevitability of the aging population mandates that all sur-geons understand geriatric specialty care issues as they pertain to perioperativecare. In 2007, more than 35% of all inpatient operations in the United Stateswere performed on adults aged 65 years and older [1]. This proportion is antic-ipated to increase in the coming decades. Optimizing the surgical community’sability to provide specialized surgical care to the older adult is vital.

Geriatric patients have unique physiologic changes compared with theiryounger adult counterparts. Frailty is the term that describes physiologiccompromise unique to older adults. Frailty is defined as a state of reducedphysiologic reserve associated with increased susceptibility to disability [2].Frailty is a unique phenotype that by definition confers increased risk ofadverse health care outcomes for community-dwelling older adults [3,4] as

Disclosures: The authors have no conflicts of interest to declare.

*Corresponding author. E-mail address: [email protected]

0065-3411/14/$ – see front matterhttp://dx.doi.org/10.1016/j.yasu.2014.05.013 Published by Elsevier Inc.

mailto:[email protected]

http://dx.doi.org/10.1016/j.yasu.2014.05.013

http://dx.doi.org/10.1016/j.yasu.2014.05.013

2 TOWNSEND & ROBINSON

well as older adults undergoing an operation [5–7]. Frailty has been promotedas a simple, brief assessment tool that can be completed before surgery with thegoal of improving the surgeon’s ability to accurately counsel older adults andtheir families about anticipated postoperative outcomes [8]. The wide adoptionof a frailty assessment in the preoperative setting has not occurred, perhapsbecause of the time required to complete a frailty assessment.

Walking speed, or gait speed, is a test that is commonly used by geriatriciansas a measure of physiologic compromise in older adults [9]. The measurementof walking speed is typically accomplished by timing a patient over a short dis-tance (eg, 5 m [15 feet]). Slower walking speed in community-dwelling olderadults has been closely related to mortality [10] and other adverse healthcare outcomes such as need for hospitalization or need for transition fromhome to an institutional care facility [11–14]. Slower walking speed as astand-alone measure was recently related to adverse postoperative outcomesin patients undergoing elective operations [15,16]. This article discusses the ex-isting evidence to determine whether measurement of walking speed alone (re-placing a multidimensional frailty assessment) can forecast postoperative risk ofmorbidity and mortality in older adults.

IMPORTANCE OF PREOPERATIVE RISK ASSESSMENT FOR THEOLDER ADULTPreoperative counseling about anticipated outcomes allows patients and theirfamilies to make informed decisions about whether or not to undergo an oper-ation. Explaining to patients and their families what outcomes they can antic-ipate following an operation is an important component of the preoperativeconsultation visit. In deciding whether to go through with a major electiveoperation, older adults often prioritize patient-centered outcomes such asfunctional independence, cognition, and physiologic health (multidimensionalentities sometimes termed health-related quality of life) rather than the standard30-day morbidity/mortality outcomes most commonly used for preoperativecounseling. For example, the chances that the patient will be left functionallydependent and therefore require institutional care in a nursing home followinghospital discharge is a piece of information that affects older adults’ willingnessto proceed with major operative interventions. Older adults living in the com-munity (ie, living independently) express that the maintenance of functional in-dependence is the most important health outcome when they prioritize theirhealth decisions [17].

HISTORY OF PREOPERATIVE RISK ASSESSMENTChoices of medical and surgical treatment are made with uncertainty of theoutcomes to be expected. Surgeons historically categorized risk as either highor low by the judgment of an authoritative physician [18]. This initial guess-work to frame a patient’s risk evolved to a more quantitative approach thatused statistical methodologies to create risk indices forecasting the occurrenceof specific postoperative complications [19]. The most widely recognized article

3WALKING SPEED AND POSTOPERATIVE MORBIDITY

marking this transition from a purely qualitative assessment to a quantitativeassessment is Goldman and colleagues’ [20] cardiac risk index published in1977 in the New England Journal of Medicine. This landmark article is the foun-dation of the current American Heart Association’s statement of the preopera-tive cardiovascular evaluation before noncardiac operations, which provides apathway for cardiac evaluation and forecasts the probability of adverse cardiacevents [21]. In addition to this algorithm for preoperative evaluation ofischemic heart disease, similar practice strategies exist to evaluate postoperativepulmonary, renal, and hepatic risk. The major limitation of these strategies isthat they limit their risk prediction to complications within a single organ sys-tem and not more globally to the patient. Clinicians have always recognizedthese limitations and, from a practical standpoint, they used these risk assess-ment indices in conjunction with their own qualitative visual test to determinewhich patients were at high risk for more global poor surgical outcomesbeyond that organ system.

Frailty assessment recently emerged as a powerful method for forecastingadverse global postoperative events in older adults [5–7]. Frailty indices allowthe quantification of the qualitative visual test that surgeons have always reliedon to help guide their surgical decision making. In a sense, the past few years ofpublished literature on frailty in surgical patients have brought the history ofpreoperative risk assessment full circle to where it began. Preoperative frailtyassessment quantifies what the authoritative physician was using qualitativelyto assign perioperative risk.

PREOPERATIVE ASSESSMENT: FRAILTYFrailty assessment is recommended as standard of care for the preoperative riskassessment of the older adult by a joint best practices statement by the Amer-ican College of Surgeons and the American Geriatric Society [22]. A frailtyassessment captures multidimensional information that subsequently quantifiesglobal surgical risk in older adults. The concept of frailty helps define a pa-tient’s physiologic or biological age; a factor that is often distinct from the pa-tient’s chronologic age [23]. Frailty is characterized by the presence of declineor loss of function in multiple systems that is continuous in ts trajectory. It isalso characterized by the risk of inability to tolerate physical stressors that isindependent of, but frequently associated with, increased disability and comor-bidities that are also increasingly present in the geriatric population [3,23]. Thefrail elderly, by definition, are at higher risk for poor health care outcomes andmortality compared with their nonfrail counterparts.

Two strategies for preoperative quantification of frailty are recommended bythe American College of Surgeons’ best practices statement [22]: the multidi-mensional frailty assessment and the phenotypic frailty assessment.

The multidimensional assessment of the frail older adult before surgery issimple, brief, and effective at predicting which older adults will have a pooroutcome [5,24]. To accomplish a multidimensional frailty assessment, baselinefrailty characteristics are measured before surgery. In one of the described


frailty assessments [24], we described quantification of 7 frailty characteristics:slow mobility (timed up-and-go score [25]), dependent function (assessment ofdependence in one or more activities of daily living), impaired cognition (as-sessed by the Mini-Cog test [26]), high chronic disease burden (quantified byboth Charlson Index score [27] and anemia of chronic disease), poor nutrition(hypoalbuminemia), and the presence of a geriatric syndrome [28] (one ormore falls in the 6 months before the operation). Patients were assigned 1 pointfor each of the 7 frailty characteristics that were found to be abnormal on pre-operative assessment. Scores range from 0 (not frail: none of the frailty charac-teristics were present) to 7 (highly frail: all 7 of the frailty characteristics werepositive). Patients were defined as frail if 4 or more of the frailty characteristicswere positive in the preoperative clinic. When comparing frail and nonfrail pa-tients aged 65 years and older undergoing an inpatient operation, frail patientswere more likely to have one or more postoperative complications (56% vs17%; P<.001), to need discharge to an institutional care facility rather thanhome (47% vs 3%; P<.001), and to require readmission in the first 30 daysfollowing discharge (28% vs 7%; P ¼ .014) [8,24]. In addition, 4 or moreabnormal frailty characteristics predicted 6-month postoperative mortalitywith 81% sensitivity and 86% specificity [5]. Multiple multidimensional preop-erative frailty assessment methods have been used effectively in the preopera-tive setting to forecast poor postoperative outcomes [6,29,30].

The second method of defining frailty is termed phenotypic frailty, whichwas initially described by Fried and colleagues [3]. Phenotypic frailty is definedby measuring decline in 5 domains: shrinking (unintentional weight loss �4.5kg [10 pounds] in the last year), decreased grip strength measured by hand-helddynamometer, exhaustion (2-question test), low physical activity (measured bythe Minnesota Leisure Time Activities Questionnaire), and slow walking speedover 5 m (15 feet) [7]. Patients were defined as nonfrail (0 or 1 abnormal do-mains), intermediate frail (2 or 3 abnormal domains), or frail (4 or 5 abnormaldomains). Complications occurred after major procedures in 19.5% of nonfrail,33.7% of intermediate frail, and 43.5% of frail patients [7]. Adjusted risk anal-ysis found the frail patient odds ratio for complications was 2.54 (95% confi-dence interval, 1.12–5.77) and the investigators concluded that frailty was anindependent predictor for surgical complications [7].

WALKING SPEED ALONE REFLECTS MULTIDIMENSIONALFRAILTYWalking speed has emerged as a simple and rapid tool that can reliably identifyfrailty in the geriatric population [31]. Measurement of walking speed is often acomponent of the multidimensional frailty assessment. The literature is full ofevidence to suggest that slower walking speed alone is a good surrogate mea-sure for other characteristics of frail older adults; including impaired cognition,the presence of geriatric syndromes, and functional dependence (Table 1).

Slow walking speed has been closely related to impaired cognition in olderadults. Impaired cognition is poor mental or brain function, and represents

Table 1Evidence that slow walking speed alone is representative of the multidimensional frail olderadult

Slow walking speed isrelated to: Evidence

(1) Impaired cognition 1. Poorer memory and executive function in slow walking olderadults [32]

2. Slow walking speed predicts future cognitive decline [33](2) Future falling episodes 1. Slower walking speed forecasted patients who fell in sub-

sequent year [12]2. Walking slowly predicted older adults with recurrent falls in

the next year [11](3) Development of

functionaldependence

1. Slow timed up-and-go score reflected subsequent difficultieswith activities of daily living at 6, 12, and 18 mo [37]

2. Slow timed up-and-go forecasted new difficulties withactivities of daily living in subsequent 12 mo [12]

(4) Decreased survival 1. Slower walking speed in increments of 0.1 m/s is associ-ated with decreased survival [10]

2. Slower walking speed related to decreased survival [40](5) The presence of frailty 1. Slow timed up-and-go is both sensitive and specific for

identifying older adults with frailty [31]


dysfunction in memory, intellectual ability, calculation, and executive function.Impaired cognition in the elderly is represented by a spectrum of diseaseranging from mild cognitive impairment to dementia. There are multiplestudies that relate slower walking speed to impaired cognition. Donoghueand colleagues [32] found that a slow timed up-and-go test was associatedwith poor memory, reduced executive function, and slower cognitive process-ing speed. In a separate study of healthy 80-year-olds, Katsumata and col-leagues [33] found that a slow timed up-and-go test predicted which patientswould subsequently develop global impaired cognitive function. The investiga-tors suggested that measuring timed up-and-go may play a role in the earlydetection of cognitive impairment. Other studies confirm the relationship ofslow walking speed and poor cognition [34,35].

Slow walking speed can be used to forecast the occurrence of future geriatricsyndromes, particularly falls. A geriatric syndrome is a ‘‘multifactorial healthcondition that occur[s] when the accumulated impairments in multiple systemsrender [older] persons vulnerable to situational challenges’’ [36]. In short, geri-atric syndromes are clinical symptoms that represent the frail older adult [28].Falling is a geriatric syndrome and slow walking speed has been associated withboth falls and recurrent falls in older adults. Viccaro and colleagues [12] foundthat both slower gait speed and slower timed up-and-go forecasted increasedfalls over 1 year in adults aged 65 years and older. Lin and colleagues [11] stud-ied 1200 community-dwelling older adults and found that slower timed up-and-go (17 seconds in patient who subsequently fell and 13 seconds in patient whodid not fall) was related to recurrent falls in the subsequent 12 months.


Slower walking speed has been related to the occurrence of functional depen-dence in older adults. Functional assessment of older adults measures whetherthey are able to perform their activities of daily living (bathing, dressing, toilet-ing, transferring, continence, and feeding); functional activities that allow a per-son to live independently at home. Wennie Huang and colleagues [37] foundthat a slow timed up-and-go forecasted future difficulties with activities of dailyliving at 6, 12, and 18 months. Multiple studies have found that a slow timedup-and-go and slow gait speed both predict increasing difficulties with activitiesof daily living in community-dwelling older adults over the following 12 to18 months [12,37].

In addition, slower walking speed has been associated with the presence ofphenotypic frailty. Savva and colleagues [31] studied 1814 community-dwelling older adults to determine whether slow walking speed alone diag-nosed phenotypic frailty. A slow timed up-and-go score was both sensitiveand specific for the diagnosis of phenotypic frailty, and they concluded thattimed up-and-go and gait speed could be used to screen older adults for frailty.

Walking speed is a component of almost every type of frailty measurement.The benefit of using walking speed as a single measure to define the frail patientis that full measurement of phenotypic frailty is time consuming and oftenimpractical. Limiting the evaluation of frailty to measuring walking speedwould likely gain wider acceptance for implementation because of its simplicityand brevity.

SLOW WALKING SPEED PREDICTS ADVERSE HEALTHOUTCOMESSlow walking speed forecasts a variety of adverse health care outcomes thatare associated with the frail older adult. Slow walking speed is so closely relatedto adverse health outcomes that it has been recommended as the sixth vitalsign for geriatric patients [38,39]. Clinical outcomes of interest to bothcommunity-dwelling and hospitalized older adults include falls, hospitalization,posthospitalization discharge to institutions, and mortality. The timed up-and-go identifies patients who are at higher risk of a subsequent fall, of developingdependence in one or more activities of daily living, and for global healthdecline [11–14].

There is a well-established relationship between slow walking speed andincreased mortality in community-dwelling older adults (Fig. 1). Laukkanenand colleagues [40] measured walking speed in 466 adults aged 75 years andolder. Slow walking speed over 10 m was associated with an increased riskof death (odds ratio, 1.98; 95% confidence interval, 1.18–3.34). Studenskiand colleagues [10] performed a pooled analysis of 9 studies on walking speedand mortality and concluded that slower walking speed was associated withdecreased survival in decreasing increments of 0.1 m/s. Examples of differingsurvival based on walking speed from this study include the median survivalfor 75-year-olds who walk fast (1.6 m/s) is 18 years for men and 23 years forwomen, compared with 75 year olds who walk slow (0.6 m/s) whose survival

Slow

Fast

Characteristic Fast and Slow Walking Patterns Median Survival (years)Walking SpeedSlow (0.6m/s)

Intermediate(1.1m/s)

Fast (1.6m/s)

65 years

712

75 years

1218

85 years

47

Gender

MaleFemale

MaleFemale

MaleFemale

1216

2028

79

1823

3241

1113

Fig. 1. Walking speed and mortality. Characteristic patterns of fast and slow walking speedsare depicted adjacent to median survival (in years) associated with male and female walkingspeed at ages 65, 75, and 85 years. (Data from Studenski S, Perera S, Patel K, et al. Gaitspeed and survival in older adults. JAMA 2011;305(1):50–8.)


is 7 years for men and 12 years for women [10]. Slow walking speed has aproven, strong association with increased risk of mortality across a variety ofelderly populations.

Slow walking speed is an excellent predictor of older adults who will subse-quently have a fall. Rothman and colleagues [41] found that slower walkingspeed was related to the occurrence of an injurious fall (odds ratio, 2.19;95% confidence interval, 1.33–3.60). This finding is similar to those of otherstudies that have related falling to slow walking speeds [11,12].

Slow walking speed additionally predicts increased need for future hospital-ization. Bouillon and colleagues [42] showed in a cohort study of community-dwelling adults aged 55 to 79 years that slow walking speed was associatedwith a 14.2% chance of hospitalization in 3 years compared with 8.5% of adultswith fast walking speed (P<.0001), which translated to a hazard ratio for hos-pitalization based on walking speed of 1.6; a ratio that was greater than anyother single measurement performed as part of the phenotypic frailty examina-tion. In addition to hospitalization, slow walking speed is incrementally predic-tive of recurrent cardiovascular events. Matsuzawa and colleagues [43] studied472 patients with history of myocardial infarction. The fastest tertile of patientshad only a 3.2% risk of recurrent cardiovascular event, whereas the slowest ter-tile had a 10-times higher risk (36.7%).

WALKING SPEED PREDICTS ADVERSE SURGICAL OUTCOMESRecent studies have directly related slow walking speed to adverse surgical out-comes (Table 2) [15,16]. Our group [15] studied the relationship of a preoper-ative timed up-and-go test to postoperative outcomes in older adultsundergoing operations across surgical specialties. In a prospective cohort studyof adults aged 65 years or older, patients were grouped by their walking speedmeasured by performance on the timed up-and-go test: fast group(�10 seconds), intermediate group (11–14 seconds), and slow group(�15 seconds). Postoperative outcomes were then compared in the 3 groupsof patients. One or more postoperative complications (defined in the groupsof cardiac, pulmonary, renal, thromboembolic, infectious, and reoperation)were more likely to occur in the slow group compared with the intermediateand fast groups. Slower timed up-and-go was associated with increased

Table 2Evidence that slow walking speed is related to adverse surgical outcomes

Slow walking speed defined as: Anticipated surgical outcomes of slow walking group:

(1) Timed up-and-go �15 s[15] Following both colorectal and cardiac operation, slowtimed up-and-go was associated with:

� Increased 1-y mortality� Increase of one or more serious complications� Increased need for discharge to an institutional carefacility

� Increased 30-d readmission rate(2) 5 m walking speed �6 s [16] Following cardiac operations, slow walking speed was

associated with:� Increased composite end point of mortality andserious complication


postoperative complications following colorectal (fast 13%, intermediate 29%,and slow 77%; P<.001) and cardiac (fast 11%, intermediate 26%, and slow52%; P<.001) operations [15]. In addition to examining complication ratesbased on preoperative walking speed, this study quantified 1-year postoperativemortality based on preoperative walking speed. Slower timed up-and-go wasassociated with increased 1-year mortality following both colorectal (fast 3%,intermediate 10%, and slow 31%; P ¼ .006) and cardiac (fast 2%, intermediate3%, and slow 12%; P ¼ .039) operations [15].

The second study in a surgical population, by Afilalo and colleagues [16], as-sessed whether walking speed was a useful predictor of mortality and majormorbidity in elderly patients after cardiac surgery. This multicenter prospectivecohort trial enrolled patients aged 70 years and older undergoing heart valvereplacement or coronary artery bypass operations. Slow walking speed wasdefined as taking 6 seconds or longer to walk 5 m. Of the 131 participants,46% were classified as slow based on their preoperative 5-m walking time.Slower walking speed was significantly related to the occurrence of the compos-ite end point of major postoperative morbidity or mortality (odds ratio, 3.05;95% confidence interval, 1.23–7.54). The study concluded that measuringwalking speed was a simple and effective test to identify older adults at higherrisk of major morbidity and mortality after cardiac operations.

Both of these studies show that the timed up-and-go or walking speed can beused as a single preoperative risk screening tool that can accurately identifyolder adults at higher risk of postoperative complications and death. As a pre-operative risk assessment tool, measuring walking speed could serve to counselolder adults and their families about anticipated outcomes and aid them duringthe preoperative decision-making process.

PRACTICAL WAYS TO MEASURE WALKING SPEEDCommon ways to measure walking speed in older adults is to perform thetimed up-and-go test or measuring walking time over a specified distance.The timed up-and-go has become a well-recognized and long-standing test of

180 degree turn

Start

End

Fig. 2. Timed up-and-go test. To complete the timed up-and-go, the clinician starts timing pa-tients as they rise from a chair, walk 3 m (10 feet) at a usual pace, and return to the chair. Thetimer is stopped when the patient is seated back in the chair. (Data from Podsiadlo D, Richard-son S. The timed ‘‘Up & Go’’: a test of basic functional mobility for frail elderly persons. J AmGeriatr Soc 1991;39(2):142–8.)


functional mobility [25]. To complete the timed up-and-go, the clinician timesthe patient rising from a chair, walking 3 m (10 feet), and returning to the chair,and timing ends when the patient is seated (Fig. 2). This test combines walkingspeed with the task of rising from a chair, which requires lower extremitystrength. Requiring lower extremity strength in addition to walking speedmay be of particular importance to the postoperative patient in whom transi-tions out of bed for postoperative mobilization are essential. The other methodto measure walking or gait speed is to time a patient walking over a measuredcourse (Fig. 3). Common distances used to time the usual pace of walking are5 m [16] and 4.5 m (15 feet) [7]. Patients are allowed to start walking before thestart of the timed test, usually by allowing them a 2-m acceleration zone beforestarting the timed test.

FUTURE DIRECTIONSCurrent evidence suggests that improving a patient’s walking speed is associ-ated with increased survival in community-dwelling older adults [39]. This sug-gestion leads to the question of whether improvement of a patient’s walkingspeed before an elective operation might improve the surgical outcome. At-tempting to improve a patient’s physical or physiologic reserves before a healthcare intervention is termed prehabilitation. At present, there is evidence thatpreoperative inspiratory muscle training before cardiac operations decreases

~2m ~2m5m or 15ft

Acceleration Phase Deceleration Phase

Timed Phase

Fig. 3. Walking speed test. To complete the walking speed test, the clinician times patientswalking over a measured course. Patients are timed walking at a usual pace over a 5-m(15-foot) course. Patients are allowed a 2-m acceleration zone before starting the timed testand a 2-m deceleration zone at the end of the timed walk. ft, feet.


both pulmonary complications and length of hospital stay [44]. In the literatureon preoperative physical therapy and surgical outcomes, most studies includepatients undergoing hip and knee operations who have slow mobility becauseof musculoskeletal pain. Older adults who are physiologically frail have notbeen included in these preintervention physical therapy trials. There is the po-tential for future studies specifically to provide preoperative physical therapy toslow-walking, frail older adults with the goal of improving not only theirwalking speed and mobility but also with the potential to improve their surgicaloutcomes.

References
[1] Hall MJ, DeFrances CJ,Williams SN, et al. National Hospital Discharge Survey: 2007 sum-
mary. Natl Health Stat Report 2010;(29):1–20, 24.[2] Buchner DM, Wagner EH. Preventing frail health. Clin Geriatr Med 1992;8(1):1–17.[3] Fried LP, Tangen CM, Walston J, et al. Frailty in older adults: evidence for a phenotype.

J Gerontol A Biol Sci Med Sci 2001;56(3):M146–56.[4] Rockwood K, Song X, MacKnight C, et al. A global clinical measure of fitness and frailty in

elderly people. CMAJ 2005;173(5):489–95.[5] Robinson TN, Eiseman B,Wallace JI, et al. Redefining geriatric preoperative assessment us-

ing frailty, disability and co-morbidity. Ann Surg 2009;250(3):449–55.[6] Dasgupta M, Rolfson DB, Stolee P, et al. Frailty is associated with postoperative complica-

tions in older adults with medical problems. Arch Gerontol Geriatr 2009;48(1):78–83.[7] Makary MA, Segev DL, Pronovost PJ, et al. Frailty as a predictor of surgical outcomes in

older patients. J Am Coll Surg 2010;210(6):901–8.[8] Robinson TN, Wu DS, Pointer L, et al. Simple frailty score predicts postoperative complica-

tions across surgical specialties. Am J Surg 2013;206(4):544–50.[9] Woo J, Ho SC, Yu AL. Walking speed and stride length predicts 36 months dependency,

mortality, and institutionalization in Chinese aged 70 and older. J Am Geriatr Soc1999;47(10):1257–60.

[10] Studenski S, Perera S, Patel K, et al. Gait speed and survival in older adults. JAMA2011;305(1):50–8.

[11] LinMR, HwangHF, HuMH, et al. Psychometric comparisons of the timed up and go, one-legstand, functional reach, and Tinetti balance measures in community-dwelling older people.J Am Geriatr Soc 2004;52(8):1343–8.

[12] Viccaro LJ, Perera S, Studenski SA. Is timed up and go better than gait speed in predictinghealth, function, and falls in older adults? J Am Geriatr Soc 2011;59(5):887–92.

[13] Shumway-Cook A, Brauer S,Woollacott M. Predicting the probability for falls in community-dwelling older adults using the Timed Up & Go Test. Phys Ther 2000;80(9):896–903.

[14] BeauchetO, Fantino B, Allali G, et al. TimedUp andGo test and risk of falls in older adults: asystematic review. J Nutr Health Aging 2011;15(10):933–8.

[15] Robinson TN,Wu DS, Sauaia A, et al. Slower walking speed forecasts increased postoper-ative morbidity and 1-year mortality across surgical specialties. Ann Surg 2013;258(4):582–90.

[16] Afilalo J, Eisenberg MJ, Morin JF, et al. Gait speed as an incremental predictor of mortalityand major morbidity in elderly patients undergoing cardiac surgery. J Am Coll Cardiol2010;56(20):1668–76.

[17] Fried TR, Tinetti ME, Iannone L, et al. Health outcome prioritization as a tool for decisionmaking among older persons with multiple chronic conditions. Arch Intern Med2011;171(20):1854–6.

[18] BergM. Turning a practice into a science: reconceptualizing postwar medical practice. SocStud Sci 1995;25(3):437–76.

http://refhub.elsevier.com/S0065-3411(14)00014-1/sref1









































[19] NeumanMD, Bosk CL.What we talk about when we talk about risk: refining surgery’s haz-ards in medical thought. Milbank Q 2012;90(1):135–59.

[20] Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial index of cardiac risk in noncar-diac surgical procedures. N Engl J Med 1977;297(16):845–50.

[21] Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 2007 guidelines on perioperativecardiovascular evaluation and care for noncardiac surgery: executive summary: a reportof the American College of Cardiology/American Heart Association Task Force on PracticeGuidelines. Circulation 2007;116(17):1971–96.

[22] Chow WB, Rosenthal RA, Merkow RP, et al. Optimal preoperative assessment of the geri-atric surgical patient: a best practices guideline from the AmericanCollege of SurgeonsNa-tional Surgical Quality Improvement Program and the American Geriatrics Society. J AmColl Surg 2012;215(4):453–66.

[23] Mitnitski AB, Mogilner AJ, Rockwood K. Accumulation of deficits as a proxy measure of ag-ing. ScientificWorldJournal 2001;1:323–36.

[24] Robinson TN, Wallace JI, Wu DS, et al. Accumulated frailty characteristics predict postop-erative discharge institutionalization in the geriatric patient. J Am Coll Surg 2011;213(1):37–42 [discussion: 42–4].

[25] Podsiadlo D, Richardson S. The timed ‘‘Up &Go’’: a test of basic functional mobility for frailelderly persons. J Am Geriatr Soc 1991;39(2):142–8.

[26] Borson S, Scanlan JM, Chen P, et al. The Mini-Cog as a screen for dementia: validation in apopulation-based sample. J Am Geriatr Soc 2003;51(10):1451–4.

[27] Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidityin longitudinal studies: development and validation. J Chronic Dis 1987;40(5):373–83.

[28] Inouye SK, Studenski S, Tinetti ME, et al. Geriatric syndromes: clinical, research, and policyimplications of a core geriatric concept. J Am Geriatr Soc 2007;55(5):780–91.

[29] Kristjansson SR, Nesbakken A, JordhoyMS, et al. Comprehensive geriatric assessment canpredict complications in elderly patients after elective surgery for colorectal cancer: a pro-spective observational cohort study. Crit Rev Oncol Hematol 2010;76(3):208–17.

[30] Lee DH, Buth KJ, Martin BJ, et al. Frail patients are at increased risk for mortality and pro-longed institutional care after cardiac surgery. Circulation 2011;121(8):973–8.

[31] SavvaGM, DonoghueOA, Horgan F, et al. Using timed up-and-go to identify frail membersof the older population. J Gerontol A Biol Sci Med Sci 2013;68(4):441–6.

[32] Donoghue OA, Horgan NF, Savva GM, et al. Association between timed up-and-go andmemory, executive function, and processing speed. J AmGeriatr Soc 2012;60(9):1681–6.

[33] Katsumata Y, Todoriki H, Yasura S, et al. Timed up and go test predicts cognitive decline inhealthy adults aged 80 and older in Okinawa: Keys to Optimal Cognitive Aging (KOCOA)Project. J Am Geriatr Soc 2011;59(11):2188–9.

[34] Eggermont LH, Gavett BE, Volkers KM, et al. Lower-extremity function in cognitively healthyaging, mild cognitive impairment, and Alzheimer’s disease. Arch Phys Med Rehabil2010;91(4):584–8.

[35] Pettersson AF, Olsson E,Wahlund LO. Motor function in subjects with mild cognitive impair-ment and early Alzheimer’s disease. Dement Geriatr Cogn Disord 2005;19(5-6):299–304.

[36] Tinetti ME, Inouye SK, Gill TM, et al. Shared risk factors for falls, incontinence, and func-tional dependence. Unifying the approach to geriatric syndromes. JAMA 1995;273(17):1348–53.

[37] WennieHuangWN, Perera S, VanSwearingen J, et al. Performancemeasures predict onsetof activity of daily living difficulty in community-dwelling older adults. J Am Geriatr Soc2010;58(5):844–52.

[38] Fritz S, Lusardi M. White paper: ‘‘walking speed: the sixth vital sign’’. J Geriatr Phys Ther2009;32(2):46–9.

[39] Hardy SE, Perera S, Roumani YF, et al. Improvement in usual gait speed predicts better sur-vival in older adults. J Am Geriatr Soc 2007;55(11):1727–34.























































[40] Laukkanen P, Heikkinen E, Kauppinen M. Muscle strength and mobility as predictors of sur-vival in 75-84-year-old people. Age Ageing 1995;24(6):468–73.

[41] Rothman MD, Leo-Summers L, Gill TM. Prognostic significance of potential frailty criteria.J Am Geriatr Soc 2008;56(12):2211–6.

[42] Bouillon K, Sabia S, Jokela M, et al. Validating a widely used measure of frailty: are all sub-components necessary? Evidence from the Whitehall II cohort study. Age (Dordr)2013;35(4):1457–65.

[43] MatsuzawaY, KonishiM, Akiyama E, et al. Association between gait speed as ameasure offrailty and risk of cardiovascular events after myocardial infarction. J Am Coll Cardiol2013;61(19):1964–72.

[44] Hulzebos EH, Helders PJ, Favie NJ, et al. Preoperative intensive inspiratory muscle trainingto prevent postoperative pulmonary complications in high-risk patients undergoing CABGsurgery: a randomized clinical trial. JAMA 2006;296(15):1851–7.














does walking speed predict postoperative morbidity?

Documents