Preoperative frailty based on laboratory data and postoperative health outcomes in patient
Highlights
+ High frailty Tevels are associated with adverse health outcomes after CABG surgery:
+ Frail patients are thigh risk of postoperative readmission and extended hospital stays
+ Lab-based frailty could capture a need for thorough frailty assessments for CAG patients
Abstract
Background
Fraityis associated with adverse surgial ontoomes. Patents with cardiovascular diseases have many'rsk factors of fait; thus, preoperative frailty evaluatio
need foran accurate ut simple Failty sereening for patents awaiting CABG surgery.
Objectives
This retrospective study aimed to determine t
‘Methods
‘We evaluated 896 patients who underwent on-pump or off-pump CABG surgery between August 1,2015 and July 31,2020 at tertiary hospital. The frailty in
association between laboratory based-frailty and patient health outcomes after CABG surgery
Results,
The patients were divided into three groups according to their preoperative FI-LAB level as low (FI-LAB <0.25, 28.0%), moderate (FI-LAB 20.25 to <0.4,548
ays (p ».008), respectively, in the high frailty group than those in the low frailty group. The od ratio for 30-day readmission was also 2.88 times higher in t
Conclusion
-Ahigh preoperative FLLAB score indeates increasing rss of adverse postoperalive outcomes among CABG surgery patients FL-LAB has potenti strengths
Abbreviations and Acronyms
APACHE
‘Acute Physiology and Chronic Health Bvaluation Il score
BMI
body mass index
cae
coronary artery bypass graft
cD
cardiovascular diseasesBuroSCORE
European System for Cardiac Operative Risk Evaluation
FELAB
frailty indexlaboratory
cu
LVEF
Jef ventricular ejection fraction
opcaB
‘off-pump CABG
sts
Society of Thoracie Surgeons
woct
‘weighted Cherlson comorbidity index
Introduction
‘The inereasing numberof elderly surgery patients emphasizes the importance ofthe influence of preoperative frailty on postoperative care. Frit is defined a
addition, potential risk factors, such as geneticjepigenetc factors o environmental/lifesiyle stressors, might facilitate its pathophysiology. * Fralty is assocz
discuss surgery’ risks wit patients, their families, and healthcare professionals.
Patients with cardiovascular diseases (CVs) have risk factors of frailty suchas deereased physical activity and diabetes ® 19 addition, common pathological
(ovDs © and 50% among older patients undergoing elective surgery, ” Hence, preoperative frailty evaluation is important for prediction of their isk of comp]
various approaches curently used, °° The chee ofa rally assessment tool depends on the availabilty of information, the reason forthe assessment, the
Previous studies onthe effet frailty onthe health outcomes of patients undergoing cardiovascular surgery have used varying fal assssment tools. Most
functional desine after cardiac surgery. * Although many studies have suggested the adverse effects of fait on outcomes after cardiac surgery, ® "8 eve
‘Generally a comprehensive frailty assessment is regarded asthe best approach to detect ail patients because ity is related to snuifactorial isk factors, in
Recently,
results and vital signs; thus, itcould bea more objective modality for assessing frailty as compared toa judgment-based evaluation. "¥ Moreover, it would be
routine clinical data-based frailty assessments have been suggested as pragmatic approsch for screening fail patients in ate settings. 7 Among,
Preoperative frailty assessment for cardiac surgery patients shouldbe quik, simple, highly accurate, and generalizable to these patents.
‘within 30 days in patients undergoing coronary artery bypass graft (CABG) surgery.
“Therefore, this tu
Methods
Study design and patients
‘This retrospective study was approved by the institutional review board of XX University Health System (seference no. ¥-2020-0100). The investigation conto
the analysis.
‘The participants were paients aged >19 years who undeswent CABG surgery between August 1, 2015 and July 81,2020 ata tertiary hospital. Patients who un
‘the FI-LAB items were missing were also excluded. % Tn total, 896 patients were include in the study (Eig 10004)Fat
{CARS cornary artery bypass gral, FAB: alt ndexlaboraony
Frailty measurement
rally was assessed by the FI-LAB scove suggested by Blodget etal. 2 This version of FI-LAB has not been validated for CABG patients yet, but it was alread
vital signs, and any value outside the reference range is considered a deficit. The frailty index was a continuous score ranging from Oto 1 and calculated by the
laboratory results before the surgery and the frst measured vital signs at hospitalization for surgery. Thus, the frailty level vas messured using data obtained |
we excuded these two variables from the analysis. We required atleast 70% of items (28 out of 32 items) to measure a valid fay index consistent with previ
FFI-LAR determines the level of fraity as «continuum between robust and fal. Therefore, it was hard to divide our population into fal patients or non-trailp
moderate frailty group, FI-LAB 20.25 to <0.4; and high frailty group, FI-LAB >0.4
Other variables
(ther variables included the patent's age sex, body mass index (BMD, history of smoking or aleubol consumption, numberof carent meiatons left vent
living based on the Koresn version of Maiti Harthel Index (MBI), which reflects patents’ fonctional capacity,” and the Acute Physiology and Chronic Het
{sa comorbidity index weighted by the relative risk of 1-year mortality to reflect both the number and severity of comorbid diseases, ® was calculated using
Statistical analysis
‘he patient characterises were described as either a peeentage ofthe total sample or asthe mean + standard deviation. One-way analysis of variance and Pe
‘variables, and particularly, the standard assumptions fr linear regression analysis using residual plots and Q-Q plots. Mltiple linear regression analysis was
30 day. The covariates fr the regression models were chosen based onthe univariate findings, which yielded signifcat dilferences or correlations at least ot
considered statistically significant
Results
‘The mean patient age was 66 + 9.0 years, and 78% were men, The mean BMI was 24.7 kg/m? , and 71.8% of the patients were overweight or obese, Hyperter
an off-pump rather then an on-pump CABG. The mean postoperative APACHE Il score was 28.1 + 5:7. The mean FI-LAB score was 0.3 + 0.1.28X, 54.9%, ane
Te +
Coil an domoprptic carats of th sample (= BS).
Variabies Mean #80 n(%)
‘Age (years) 660290
<40 3103)
4049 299.2)Variables Mean + SD
50-58
0-79
280
Sex
Female
Male
BMI (agi? } (n= 874) A722
Underweight (<18.5)
Normal (18.5-22.9)
‘Overweight (23-24.9}
Obese (225)
‘Smoking status
None
ex
current
Drinking status
None
&
current
FLAS 0304
Low (<0.25)
Moderate (20.25-<0.4)
High 0.4)
wee! 2asta
o
12
oa
25
Comorbidies (matple responses)
Hypertension
Diabetes
Myocardial infarction
Dystipidemia
Hear falco
Number of madicatons (n= 837) 75236
Modified Barthel Index: ons eee
Partly dependant
Independent
LVEF (%) (n= 668) 5440150
nem)
a7 (79)
387 (388)
308 (344)
28181)
197 (22.0)
699 (78.0)
18004)
228 (26.1)
252 (288)
376 (43.0)
455 (60.8)
200 (31.2)
161 (180)
508 (56.1)
9)
250 27.9)
143 (1
208 (23.0)
492 (54.9)
198 (22.1)
129,144)
4318.1)
228 (252)
110123)
617 (68.9)
498 (65.8)
314,50)
295 02.9)
99 (11.0)
576.4)
839 (93.6)Variables
Mean SD n(%)
<40 184(177)
40-49 189,178)
50-59 183 21.1)
260 378 495)
Ccardovascuar events (n= 670)
‘Angina within 2 mantns 498 (50)
iil wihin 3 weeks 218 25.1)
‘fio during hospitalization 182)
Surgery ype
case 53,69)
orcas 243 (04
Number of grats (= 883)
1 1922)
2 09 (78)
3 595 60)
4 214 (242)
5 4662)
[APACHE It score
“et: til rian; AMI: cut myocar nfetcton APACHE: Act Physilegy and Girone He
23.1257
uation: BMI body mass index: CABG: coronary artery bypass gat: Fl
‘ie 210002) shows a descriptive comparison of FI-LAB variables among the throe groups. Compared with the moderate and low frailty groups, the high fai
associated with bing rail. Readmission within 30 days was also associated with frailty level. Moreover, the postoperative length of hospital stay and length of
FHLB peramtors by lve of tity (= 896,
Variables (Unit) Low fraty (= 208) Moderate fraty (n= 492) igh fralty (= 198)
‘Albumin (ia.) 42203 atso4 a6s05
‘Alkaline phosphatase (UIL) 6044209 ear e2ta e2as645
Bicarbonate (mmol) 228420 zane23 218232
Bilirubin, total (mg/dL) osso2 oss03 oss03
Blood pressure: diastolic (mg) 75.1 £82 reas 0022166
Blood pressure: systolic (mmHg) 127.5 11.5 134.1219 1472283
Blood urea nitrogen (mala) 159244 175468 248.2146
C-reactive protein (git) 24299 50 120 1262209
Creatinine (mid) oss02 tor08 26428
Direct HDL-cholesterol (mgidl) 4289.2 4244103 387299
Folate, RBC ingim) . . .
‘Glucose (mgidL) 13352638 148.4259.0 15792725
{Glycohemagltin loves (4) ast7 73218 72216
Horaglebin (lL) 10813 130818 waste
Iron, atigorated (yall) sor s275 a262337 7052482
Lactate dehydrogenase (UL)Variables (Unit) Low frat (= 208) Moderate fray (= 492) igh frat (= 198)
Mean arterial pressure (mmiig) 92.683. 9542129 10252 18.9
Mean cell volume (L) o19437 o1s4s 919259
Phosphorus (mmol) 3eso4 e206 37210
Platelet count (1000 cellspl) 224.1 287.0 2206 263.2 21302762
Protein, total (gl) 69204 67406 62207
Pulse (beatsimin) 7282105 7462145 8162179
Pulse prossure (mmHg) 524495 5812148 87.02207
Rd coll dstibuton width (4) 128 +08 131409 1a2e18
‘Segmented neutrophils percent (%) 58.0 +83 505¢91 6354101
‘Sodium (mmol) 1404s 18, 1400227 188237
Total calcium (mgicL) e204 sor0s aes07
‘otal cholesterol (mmol) 181.0209 21242509 18402976
“Teglycoride (mala) wr2eon7 13922817 y2022617
Utie ac (mgd) 54e18 55218 5az19
‘Vitamin 612 (pgm) e1222698 © 645.9+978.7 673.22 9498
‘Vitamin 0 (ng) 248400 s9B 4118 214275
“sig variables.
Pata ate presented s the mean # SO.
{FHLAB: ray indexsaborloy, HOLhighndensty lporotans, RBC: re blod el
Coal ans demoprannie caractersts ofthe saleby F-LAB ally group (= 898)
Variables Total (n= 696) Low falty (n= 208) Moderate frat (= 492) High fally (n= 198) F ory? (p)
‘Age (years) e60r00 © 635299 eseee8 e741 £89 10.39 (<0.001) a
e1 ghana)
Underweight se@t) 240) 1123) sen 8.91(0.174) “tama
Normal 228 (26.1) 4822.5) 120 (24.8) 62 (382)
‘Overweight 252 (288) 59283) 144,298) 49262)
Obese 376 (430) 97 (47:5) 208 (43.1) 71 (38.0)
‘Smoking status
None 455 (608) 110 (52.4) 243 (49.4) 102 (51.5) 3.68 0.455)
ex 200 (912) 98 (32.0) 15992.) 55 (278)
current 161(18.0)90(14.6) 90 (18.3) 41207)
Drinking statusVariables
None
&
Current
weet
°
my
ew
25
Number of medications
ADL
Dependent
Independent
LVEF (%)
285
<5
‘Angina within 2 months
y
N
AMI within 3 weeks.
Y
N
‘Acib during hospitalization
Y
N
‘Surgery type
CABG
oPcas
APACHE It score
Length of hospital stay (ays)
Length of ICU stay (days)
Readmission within 30 days
+ Fisher's exact
+ Bonferroni post-hoc est
Total (= 896) Low tally n= 206) Moderate fraity (n= 482) High ally (n= 198) F ory? (py
503 (6.1)
143 (160)
250 (278)
2aet9
129,168)
431 (84)
226 (252)
10123)
18238
87064)
#29 (93.6)
5442150
468 (69.9)
400 (46.1)
496 60)
492 49.7)
218 25.1)
652 (74.9)
1821)
852(97.9)
£3169)
849 (04.1)
234257
113293
33237
86 (8.6)
121 6827)
mann
61 298)
ssetz
98272)
125607)
22 (0037)
345)
70234
6.29)
200 (97.1)
5842134
132 66.0)
08 (340)
90 (45.0)
110 (65.0)
26 (13.0)
174,870)
+105)
199 (9955)
1268)
194 (942)
720254
o3248
2asia
12 (6.8)
270 (64.8)
76154
146 2927)
past
671138)
256 (620)
119 262)
50 (102)
75235
2116)
465 (945)
s4a2150
260 (645)
217 (855)
227 (87.4)
282(62.6)
108 (22.5)
anq7s)
7115)
472,985)
21188)
465 (04.5)
726356
wosere
26224
47(8.6)
‘pump coronary artery bypass; wCCI: weighted Charleom comorbidity index.
112668)
43217
43217
ag222
630)
50 25.)
85 429)
51 (2838)
e239
2412.1)
174 (879)
4932180
75398)
115 (602)
125 (634)
70 98.6)
84 (44.0)
107 (56.0)
10152)
181 (04.8)
1407)
194 (28)
256460
1470135
43250
2713.8)
10.47 0.032)
112.10 (<0.001) 9b < of guanay
$78.12 (<0.001)
12.48 (<0.001) a, 6< 1 gusty
18.77 (<0.001)
20,02 (<0.001) a> b> esta
21.21 (<0.001)
17.202 (0.002)
53,671 (0.001)
1.01 (0.021) - quate
0.84 (0.728)
23,79 (<0.001), b= cana)
217 (<0.001) a, b< sn
242 (<0.001)a, b< cl ghana
7.4 (0.028)
fib tral brillation; AMI: seute myocardial infarction; APACHE: Acute Physiology end Chronic Health Evaluation; BMI: body
{In the covarate-adjusted analysis, the length of hospital stay and the length of ICU stay were longer by 2.20 days ( p ~028) and 0.89 (p=.008), respectively,
the low frailty group (able 4 o004)
Etec fal on postoperative heath atcomes.Length of hospital stay Tmt) Length of ICU stay 1 got) Readmission within 30 days 3 geamnz)
6 valve et) 8 pvawe en) OR (95% cH)
FLLAB
Low (Reference)
Modorate 036638 “oor 975 151 0.76299)
gn 220 028 069 009 2.586 (115-560)
woot
0 (Reference)
12 02 ate oot 978 1.23 (056-268)
a 170.098 08.057 077 (911.92)
2s 642