Traditional Intravenous Fluid Versus Oral Fluid Administration in Primary Total Knee Arthroplasty: A Randomized Trial

Traditional Intravenous Fluid Versus Oral Fluid Administration in Primary Total Knee Arthroplasty: A Randomized Trial

Journal Pre-proof Traditional Intravenous Fluid vs. Oral Fluid Administration in Primary Total Knee Arthroplasty: A Randomized Trial Jason M. Jennings...

433KB Sizes 0 Downloads 32 Views

Journal Pre-proof Traditional Intravenous Fluid vs. Oral Fluid Administration in Primary Total Knee Arthroplasty: A Randomized Trial Jason M. Jennings, MD, DPT, Mauricio Mejia, MD, Michael A. Williams, MD, Roseann M. Johnson, BS, Charlie C. Yang, MD, Douglas A. Dennis, MD PII:

S0883-5403(20)30066-8

DOI:

https://doi.org/10.1016/j.arth.2020.01.029

Reference:

YARTH 57737

To appear in:

The Journal of Arthroplasty

Received Date: 2 December 2019 Revised Date:

7 January 2020

Accepted Date: 14 January 2020

Please cite this article as: Jennings JM, Mejia M, Williams MA, Johnson RM, Yang CC, Dennis DA, Traditional Intravenous Fluid vs. Oral Fluid Administration in Primary Total Knee Arthroplasty: A Randomized Trial, The Journal of Arthroplasty (2020), doi: https://doi.org/10.1016/j.arth.2020.01.029. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Elsevier Inc. All rights reserved.

Traditional Intravenous Fluid vs. Oral Fluid Administration in Primary Total Knee Arthroplasty: A Randomized Trial Jason M. Jennings, MD, DPT1,2 Mauricio Mejia, MD3 Michael A. Williams, MD4 Roseann M. Johnson, BS1 Charlie C. Yang, MD1 Douglas A. Dennis, MD1,2,5,6

1

2

Colorado Joint Replacement, Porter Adventist Hospital, Denver, CO, USA Department of Mechanical and Materials Engineering, University of Denver, Denver, CO, USA 3 US Anesthesia Partners, Denver, CO, USA 4 Porter Adventist Hospital, Denver, CO 5 Department of Orthopaedics, University of Colorado School of Medicine, Denver, CO, USA 6 Department of Biomedical Engineering, University of Tennessee, Knoxville, TN, USA

Correspondence to: Jason Jennings, MD, DPT 2535 S Downing St. Suite 100 Denver, CO 80210 [email protected] The study was performed entirely at Colorado Joint Replacement, Porter Adventist Hospital. 2019 James A. Rand Young Investigator’s Award

1 2

Traditional Intravenous Fluid vs. Oral Fluid Administration in Primary Total Knee Arthroplasty: A Randomized Trial

3 4

BACKGROUND: Optimal perioperative fluid management has not been established in patients

5

undergoing orthopaedic surgical procedures. Our purpose was to investigate the effects of

6

perioperative fluid management (i.e. preoperative, intraoperative and postoperative) on patients

7

undergoing TKA.

8

METHODS: One hundred thirty patients who met inclusion criteria undergoing primary

9

unilateral TKA were prospectively randomized into traditional (TFG) vs. oral (OFG)

10

perioperative fluid management groups. The primary outcome was change in body weight

11

(BW). Secondary outcome measures included: knee motion, leg girth, bioelectrical impendence,

12

quadriceps activation, functional outcomes testing, KOOS JR, VR-12, laboratory values, vital

13

signs, patient satisfaction, pain scores, and adverse events.

14

RESULTS: The TFG had increased BW the evening of surgery (7.0±4.3 vs. 3.0±3.9,

15

p<0.0001), post-operative day (POD) #1 (9.1±4.3 vs. 4.7±3.9, p<0.0001), and POD #2 (6.2±5.0

16

vs. 4.4±4.0, p=0.032). Bioelectrical impedance showed less limb edema in the OFG (4.2±29.7

17

vs. 17.8±30.3, p<0.0001) on POD#1. Urine specific gravity differences were seen preoperatively

18

between groups (OFG, more hydrated, p=0.002). Systolic blood pressure decrease from baseline

19

was greater in the OFG upon arrival to the floor (19.4±13.5 vs. 10.6±12.8, p<0.0001) and 8

20

(23.4±13.3 vs. 17.0±12.9, p=0.006) and 16 (25.8±13.8 vs. 25.8±13.8, p=0.046) hours after floor

21

arrival. The TFG had more UOP on POD#1 (3369mL±1343mL vs. 2435mL±1151mL, p <

22

0.0001). The OFG were more likely to go home on POD #1 than the TFG (63 vs. 56, p=0.02).

pg. 1

23

CONCLUSIONS: Oral fluid intake with IVF restriction in the perioperative period after TKA

24

may offer short-term benefits with swelling and BW fluctuations. The authors continue to limit

25

perioperative IVFs and encourage patient initiated fluid intake.

26

Keywords: Perioperative Hydration; total knee arthroplasty; Intravenous Fluid; Anesthesia

pg. 2

27 28

BACKGROUND Perioperative fluid management and its implications in elective surgery are controversial,

29

with a large variability with regards to optimal management in daily practice (1,2,11–15,3–10).

30

Liberal vs. restrictive fluid perioperative management has been defined in the literature but there

31

have been differences with regards to the methodology and results in the reported studies in

32

patients undergoing major abdominal surgery. It is proposed that a liberal approach to fluid

33

management (i.e. traditional fluid management) may lead to volume overload causing a decrease

34

in pulmonary function, postoperative bowel mobility issues, and an increase in cardiac demand

35

(5,11,12). However, others have suggested a restrictive or a goal-directed fluid regimen may

36

offer no advantages (6–8) and may increase complications such as acute kidney injury (4).

37

Although fluid therapy remains a cornerstone of surgical practice, evidence-based

38

guidelines on optimal perioperative fluid management have not been established in patients

39

undergoing elective arthroplasty surgery procedures. In a small trial of patients undergoing fast-

40

track total knee arthroplasty (TKA), a liberal fluid regimen lead to a reduction in vomiting but an

41

increase in hypercoagulability without other differences noted in their recovery variables when

42

compared to a restrictive intravascular fluid protocol (10). “Goal directed” fluid therapy

43

improved patient outcomes by decreasing postoperative complications compared with traditional

44

liberal fluid management in a small cohort of patients undergoing elective total hip arthroplasty

45

(THA) (14).

46

The purpose of this prospective, randomized controlled trial was to investigate the effects

47

of two regimens of perioperative fluid management (i.e. preoperative, intraoperative and

48

postoperative), a traditional (i.e. more liberal) vs. oral (i.e. restrictive or goal-directed) directed

49

therapies. We hypothesized that the oral fluid group (OFG) (i.e. patient driven volume intake)

pg. 3

50

would maintain body weight (primary outcome) which may lead to a more normovolemic

51

physiologic state in the perioperative period. Secondary outcomes included: knee motion, leg

52

girth, bioelectrical impendence, quadriceps activation, functional outcomes testing, KOOS JR,

53

VR-12, laboratory values, vital signs, patient satisfaction, pain scores, and adverse events.

54 55 56 57

METHODS

Our institutional review board approved this study prior to initiation which was registered

58

with ClinicalTrials.org (NCT03719378). We identified patients with primary osteoarthritis

59

scheduled for unilateral primary TKA by one of four surgeons at our institution between

60

8/15/2016-7/31/2018. Patients were eligible for inclusion if they were 75 years old or younger

61

undergoing elective unilateral primary TKA for degenerative osteoarthritis. Exclusion criteria

62

included patients with volume-dependent cardiac conditions (i.e. aortic stenosis, pulmonary

63

valve stenosis, aortic insufficiency), pulmonary hypertension, cardiac dysrhythmias,

64

abnormalities in the hypothalamic-pituitary axis, stage 3 or 4 renal disease, insulin dependent

65

diabetics or diabetics with a hemoglobin A1C greater than 7, a body mass index greater than 38

66

or less than 19, current long acting narcotic medication or patients on short acting narcotic

67

medication for 3 or more months, untreated or uncontrolled GERD, preoperative anemia

68

(hemoglobin < 14 for men and < 12 for women) , patients unable to receive a spinal anesthetic

69

and patients at risk for electrolyte abnormalities (i.e. diuretics) (Figure 1). Patients older than 75

70

years old were excluded as we felt the physiologic demand in these patients could represent

71

outliers and for the purpose of this study we were attempting to capture “healthy” patients as our

72

pilot.

73 74

Patients were prospective randomly assigned into “traditional” (TFG) vs. OFG for preand post-operative fluid management at their pre-operative appointment. Randomization was pg. 4

75

performed utilizing Minitab. All patients underwent pre-operative medical screening. Pre-

76

operative laboratory values consisting of a complete blood count (CBC) and basic metabolic

77

panel (BMP) were obtained per our standard protocol.

78

All patients in this study had 1 of 13 selected anesthesiologists to assure consistency

79

with the protocol pre- and intra-operatively. All patients received a spinal anesthetic consisting

80

of 9 mg (1.2 ml) –11.25 mg (1.5ml) of spinal bupivacaine 0.75% with dextrose 8.25%. The

81

anesthesiologist also administered an ultrasound guided adductor canal block utilizing 30 mL

82

of preservative free ropivacaine 0.35%. The anesthesiologist documented the total amount of

83

vasopressors (e.g. Ephedrine and Phenylephrine) given intraoperatively. Ondansetron (4 mg)

84

was given prior to regional anesthesia and an additional dose was given at the end of the

85

operative procedure.

86

Personalized scales were given to each patient to avoid variances in weight measurements

87

during the study. TFG subjects did not eat or drink after midnight prior to their surgery. Fluid

88

initiation began intraoperatively by the anesthesiologist consisting of 2 liters of lactated Ringers

89

solution. Post-operatively 2 additional liters were administered postoperatively 500 mL in the

90

post-operative care unit and 1500mL after the patient was transitioned to the inpatient floor. The

91

amount of fluids administered in the perioperative period were based on averages given at our

92

institution to our TKA patients. These patients received a normal diet post-operatively.

93

The OFG was encouraged to drink a minimum of three, 20-ounce cups of clear liquids

94

(i.e. water, clear sports drinks, fruit juices without pulp) per day each day for the three days prior

95

to the procedure. Prior to the procedure, patients were unable to intake food or milk 8 hours

96

before surgery. Patients drank 10-ounces of clear liquids four hours prior to surgery. Intra-

97

operatively the patients received lactated Ringer intravenous fluids at a rate of 75 mL/hr. This

pg. 5

98

was discontinued when the patient began oral fluid intake post-operatively or when the total

99

amount of IV fluids given reached 500 mL. Post-operatively, beginning when the patient was

100

transferred to the inpatient floor, each patient consumed a minimum of three, 20-ounce cups of

101

liquids per day for the first three postoperative days. The amount of fluids administered to the

102

OFG was an attempt to create a “normovolemic physiologic” state based on collaboration

103

between our orthopaedic surgeons, anesthesiologist and internal medicine teams. All patients

104

otherwise continued with a normal diet post-operatively. The patients randomized to this cohort

105

kept a detailed log with regards to fluid intake.

106

All patients received 1 gram of tranexamic acid prior to incision and 1 gram prior to

107

closure. A tourniquet was not used in this study unless the surgeon preference was to use just

108

during the cementation of implants. All TKAs, regardless of group assignment, were performed

109

using a medial parapatellar approach and a gap balancing technique with substitution of the

110

posterior cruciate ligament. Before wound closure, all knees received an intraarticular

111

(synovium, periosteum, posterior capsule, and arthrotomy site) injection of a cocktail containing

112

Ropivicaine, Epinephrine, Clonidine, and Ketorolac to enhance postoperative analgesia control.

113

Anti-inflammatory medications were standardized, and aspirin was used for deep vein

114

prophylaxis unless their risk stratification suggested a reason for more potent chemoprophylaxis.

115

In these cases, Apixaban (Bristol-Myers Squibb) was utilized. Patients were provided with the

116

same standard rehabilitation protocol during their inpatient stay. Outpatient physical therapy was

117

performed per our standard protocol at our institution.

118 119

All patients received Foley catheterization to allow for an accurate measure of urine output in the perioperative phase of care. Urine specific gravity (USG) was obtained just prior to

pg. 6

120

surgery as a surrogate of hydration. USG, CBC and BMP laboratory values were again assessed

121

on post-operative day #1 and at the two-week follow-up visit.

122

Patient body weight (BW) was assessed pre-operatively and on POD#1, POD #2, 7 and

123

14 days postoperatively. An additional measurement was obtained 3 weeks postoperatively if

124

the patient’s body weight at 2 weeks had changed more than +/- 2% from their preoperative

125

value. Patients were provided their own scales to avoid potential calibration differences that may

126

affect results.

127

Isometric quadriceps muscle strength was measured during a maximum voluntary

128

isometric contraction (MVIC) using a Hand Held Dynamometer (Lafayette Instrument,

129

Lafayette, IN, USA) during preoperative, POD #1, 2 week, and 6 week assessments. Quadriceps

130

strength assessments were performed twice, and the maximum voluntary force value was

131

recorded. If maximal force differed by more than 5%, a third trial was performed.

132

Functional performance testing was assessed at 2 and 6 weeks postoperatively and

133

included the timed up and go (TUG) and the 30 second sit-to-stand tests. Knee Society Score

134

(KSS), Knee injury and Osteoarthritis Outcome Score (KOOS) and the VR-12 were collected

135

from the patients pre-operatively and at their 2 week and 6 week follow up appointments.

136

Bioelectrical impedance swelling (BIS) assessment was performed using RJL Systems

137

Quantum® to determine the difference in gross limb swelling between the operative extremities.

138

This device utilizes a four-wire measurement method, that is, two electrodes provide a current

139

and two electrodes measure the voltage drop. Lower levels of impedance indicate greater

140

swelling. Surface electrodes were placed on the lower extremity using a previously documented

141

method, by which surface electrodes are placed on the dorsal surface of the foot and anterior

pg. 7

142

surface of the thigh. BIS has previously been described to assess swelling after TKA (16–18).

143

These measurements were done pre-operatively, POD#1, and at 1 and 2 weeks post-operatively.

144

Other measures that were assessed included active knee range of motion (AROM),

145

measured in the supine position using a long-arm goniometer. Lower extremity girth measures of

146

swelling were obtained 10 cm proximal to the lateral malleolus (calf girth), at the superior pole

147

of the patella (superior patellar girth), and 10 cm proximal to the superior pole of the patella

148

(thigh girth). Pain was measured using a visual analog scale (VAS) performed at the beginning

149

of each assessment session.

150

Adverse events (wound complications, DVT, transfusions) were closely monitored and

151

documented. Particular attention was paid to adverse events that may be related to fluid

152

management (i.e. blood pressure, nausea/emesis). Lastly, major complications and reoperations

153

were recorded and reported.

154

Patients demographic data was compared between the two groups using student’s t-test

155

on continuous data (age and BMI) and chi-squared test for association on categorical variables

156

(gender) to ensure there was no statistical significance between the two groups. The primary

157

outcome (change in BW) and secondary’s outcomes were analyzed in a similar fashion. A

158

power analysis was performed prior to the initiation of the study. It was determined that 65

159

patients in the treatment and control groups were needed in order to achieve a power of 0.8.

160

RESULTS

161

Patient characteristics are presented in Table 1 and demonstrate no statistical significance

162

found between the two group’s age, gender and BMI. A total of 130 patients (65 in the TFG and

163

65 in the OFG) completed the study through the 6-week follow-up timepoint. Urine specific

164

gravity differences were seen preoperatively between groups (OFG, more hydrated, p=0.002),

pg. 8

165

Table 2. Most patients did not find the protocol difficult to follow in the TFG or OFG cohorts

166

(58 vs. 62, p=0.228). The median amount of IVF administered in the perioperative period was

167

significantly greater in the TFG vs. OFG (3895 mL vs. 350 mL p= < 0.0001).

168

Table 3 presents mean differences in the change in body weight (primary outcome)

169

through 2-weeks post-operatively. The TFG had increased BW the evening of surgery (7.0±4.3

170

vs. 3.0±3.9, p <0.0001), post-operative day (POD) #1 (9.1±4.3 vs. 4.7±3.9, p <0.0001), and POD

171

#2 (6.2±5.0 vs. 4.4±4.0, p=0.032). At the one and two-week assessments, there were no

172

significant differences noted between the groups.

173

Secondary outcomes of bioelectrical impedance showed less limb edema in the OFG

174

(4.2±29.7 vs. 17.8±30.3, p<0.0001) on POD#1 (Table 4). Leg girth at the mid-patella

175

measurement was significantly greater in the OFG at the 2-week assessment (3.0 ± 2.0 vs. 2.2 ±

176

2.3, p=0.05), Table 5. The TFG VR-12 physical score demonstrated a significant decrease from

177

their preoperative baseline compared with that of the OFG at 2-weeks (-8.3 ± 10.0 vs. -4.5 ±

178

11.2, p=0.046), Table 6. This correlated with the KSS at 2-weeks showing a significantly greater

179

decrease from baseline in the TFG (-14.6 ± 39.9 vs. -0.2 ± 36.2, p=0.034), Table 6. Systolic

180

blood pressure decrease from baseline was greater in the OFG upon arrival to the floor

181

(19.4±13.5 vs. 10.6±12.8, p<0.0001) and 8 (23.4±13.3 vs. 17.0±12.9, p=0.006) and 16

182

(25.8±13.8 vs. 20.9 ± 14.0, p=0.046) hours after floor arrival (Table 7). The TFG had more UOP

183

on POD#1 (3369mL±1343mL vs. 2435mL±1151mL, p < 0.0001). The OFG were more likely to

184

go home on POD #1 than the TFG (63 vs. 56, p=0.02).

185

Significant laboratory value differences were noted between the TFG and OFG with

186

BUN (2.6 ± 4.4 vs. 0.5 ± 4.7, p=0.01), calcium (0.9 ± 0.4 vs. 0.6 ± 0.4, p<0.0001), creatinine (0.0

187

± 0.1 vs. -0.1 ± 0.1, p<0.0001), sodium (-0.3 ± 2.3 vs. 2.5 ± 2.9, p<0.0001), on POD#1 compared

pg. 9

188

with preoperative values (Table 8). Glucose was significantly increased in the TFG vs. OFG at

189

the 2-week postoperative visit (2.5± 20.4 vs. -3.9 ± 16.4, p=0.05). No other differences were

190

noted at either timepoint between the two groups (Table 8).

191

Average self-reported VAS pain scores on POD #1 (2.6 ± 1.4 vs. 2.3 ± 1.4, p=0.324),

192

POD#2 (4.4 ± 1.5 vs. 4.5 ± 1.6, p=0.885), week #1 (4.3 ± 1.6 vs. 4.1 ± 1.6, p=0.330), week #2

193

(3.6 ± 1.8 vs. 3.4 ± 1.5, p=0.581), week #3 (3.0 ± 1.7 vs. 3.2 ± 1.7, p=0.475), week #4 (2.7 ± 1.6

194

vs. 2.6 ± 1.4, p=0.902) did not differ between the TFG and OFG. Change in AROM on POD #1

195

(-30.5 ± 18.5 vs. -26.7 ± 16.0, p=0.219), week #2 (-27.9 ± 17.5 vs. -25.5 ± 16.5, p=0.431) and

196

week #6 (-7.7 ± 17.0 vs. -4.8 ± 13.2, p=0.282) revealed no significant differences between the

197

TFG and OFG. No differences in quadriceps activation between the TFG and OFG were present

198

on POD #1 (-10.4 ± 7.6 vs. -11.2 ± 7.1, p=0.515), week #2 (-9.0 ± 6.5 vs. -9.6 ± 7.2, p=0.621)

199

and week # 6 (-3.2 ± 5.8 vs. -3.8 ± 6.5, p=0.544). The timed up and go test at week #2 (4.4 ± 4.8

200

vs. 3.3 ± 3.7, p=0.155) and week #6 (-4.7 ± 3.5 vs. -4.0 ± 3.9, p=0.283) had no differences

201

between the TFG and OFG respectively. Likewise, the change in 30 second sit to stand from

202

preoperative values revealed no differences at week #2 (-5.3 ± 4.0 vs. -4.4 ± 4.6, p=0.251) and

203

week #6 (-0.4 ± 4.1 vs. 0.3 ± 3.0, p=0.349).

204

There were no differences in post-operative nausea (p=0.717) or vomiting (p=0.753)

205

between the groups. Procedure time (74.9 ± 15.1 vs. 78.9 ± 17.9, p=0.164) (Table 1), blood loss

206

(136.2 ± 46.1 vs. 128.5 ± 47.4, p=0.350) (Table 1), and the amount of intraoperative

207

vasopressors administered (p=0.854) did not differ between the TFG and OFG.

208

There was one readmission in the TFG (acute pulmonary embolism) and none in the OFG

209

(p=0.315). No patients in either group required a transfusion. One patient in the TFG and OFG

210

had nausea on POD#1, neither patient required treatment. One patient in the OFG received a 500

pg. 10

211

mL bolus of fluid on POD#1. One patient in the TFG group required insertion of a Foley

212

catheter upon discharge secondary to urinary retention.

213

DISCUSSION

214

Despite being critically important in surgical practice, little evidence-based guidelines

215

exist in orthopaedic surgery regarding optimal perioperative fluid management. To our

216

knowledge, this is the largest study to investigate traditional fluid management which has

217

typically been the more liberal use of perioperative fluids to a more patient driven fluid

218

restrictive protocol. The OFG demonstrated a decrease in BW the first 48 hours (primary

219

outcome) after TKA. Lower leg edema, UOP and LOS all were decreased in the restricted fluid

220

group as well. Our results suggest limiting fluid intake in the perioperative period after TKA has

221

beneficial effects and can be safely done without increasing the risk of complications.

222

Goal-directed fluid and restrictive fluid management has gained widespread use in

223

elective surgery outside of orthopaedics. However, controversy regarding optimal fluid

224

management in this patient population continues (1,2,11,12,15,3–10). It has been recognized

225

that the method (IVF vs. oral) of fluid administration may play a role as well (11). Perioperative

226

IVF may exacerbate tissue injury in the setting of surgery causing accumulation in the interstitial

227

space which typically leads to unwanted edema. Our results showed had a significant decrease

228

in weight gain (primary outcome) in the early post-operative period in the OFG. This translated

229

into less lower leg edema in the operative extremity in the OFG as demonstrated by BIS

230

measurements between the two cohorts on POD#1. These findings may translate into better pain

231

management that was not captured with our current cohort and data collection points.

232

Additionally, less edema may lead to a decrease incidence of wound complications that our study

233

was not powered to detect a difference. To our knowledge, there are no studies available

pg. 11

234

assessing these outcomes (i.e. BW, lower leg edema, wound healing after TKA) for comparative

235

data. Further studies are warranted in this area to determine the true clinical significance. The

236

authors’ continue to practice with a restrictive patient driven fluid protocol but individualize this

237

based on patient specific needs in the perioperative period.

238

The results of this study suggest the perioperative complication rates were not

239

significantly different between the two cohorts. Perhaps most important is that hemodynamic

240

stability did not appear to be affected in the OFG. While there was a significant decrease in SBP

241

when the patients arrived to the orthopaedic unit, this did not translate to episodes of hypotension

242

requiring treatment intervention. The OFG was more likely to go home on POD #1. This is

243

consistent with other findings in elective abdominal surgery showing a shorter LOS in patients

244

who received less perioperative fluids (5,11). These findings may have implications not only for

245

hospital-based TKA but also for patients receiving this intervention as an out-patient procedure.

246

Previous small cohorts in TKA and THA suggest there may be some benefit to a more

247

restrictive fluid protocol in the perioperative period. These studies reported a decrease in overall

248

complications (14) and hypercoagulability (10) in patients receiving a restrictive or goal-directed

249

fluid protocol. However, more vomiting was noted in the restrictive protocol in patients

250

undergoing TKA (10). Our findings do not support the previous studies. With regards to overall

251

complications we did not see a difference in our cohorts. These differences are likely

252

multifactorial including the larger size of our cohort, differences in anesthesia practices (regional

253

vs. general) and patient population between the studies. However, the studies (including the

254

current report) are not powered to adequately detect true differences in these secondary outcomes

255

and further studies are warranted in this area to determine the most optimal fluid management

256

approach in this patient population.

pg. 12

257

Some degree of oliguria in response to the stress of surgery should be expected in

258

patients as a normal physiological response (11). This is not necessarily abnormal particularly

259

without signs of hypoperfusion. However, post-operative urinary retention (POUR) remains a

260

common complication after TJA (19,20). POUR has been shown to increase LOS and potential

261

increase in UTI (19). An increase in intraoperative fluid volume has been identified as a risk

262

factor for POUR in the TJA population (19,20). While not a primary outcome in this study, our

263

results suggest that a restricted amount of perioperative fluids can be administered in patients

264

undergoing TKA without significant consequence.

265

This study is not without limitations. While this study is powered for our primary

266

outcome, this is still a relatively small cohort. We did not assess pulmonary function or exercise

267

capacity which may be affected by fluid overload. Hormonal responses (i.e. aldosterone, anti-

268

diuretic hormone, angiotensin II) were not monitored which are the primary mediators of fluid

269

hemostasis. We excluded patients with baseline volume dependent conditions (i.e. pulmonary

270

hypertension, congestive heart failure), thus our results may not be applicable to this patient

271

population undergoing an elective primary TKA. We concede there may have been variability in

272

the amount of fluid intake in both cohorts that were not controlled for in this study.

273

Additionally, the role of no liquids prior to 8 hours of surgery has recently been challenged (11).

274

There is evidence demonstrating clear liquids 2 hours prior to surgery does not increase gastric

275

volumes and may even reduce acidity of the stomach fluids (11). Patients in our study were

276

allowed clear liquids 4 hours prior to surgery. We do not think the difference in 2 hours likely

277

would have changed our results but certainly is a variable that we did not account for that is in

278

our current practice. Lastly, all of the patients in this study received spinal anesthesia, therefore,

279

these fluid requirements and findings may differ in patients undergoing TKA with general

pg. 13

280

anesthesia. Despite these limitations we feel that the restrictive fluid protocol appears to be safe

281

and appropriate offering advantages in the early perioperative period.

282

In this prospective randomized trial, the use of a self-driven more restricted fluid protocol

283

showed less weight gain, edema and UOP in the early perioperative period. Additionally,

284

patients in this cohort had a shorter length of hospitalization. There was not an increase rate of

285

complication in the restrictive fluid group indicating that this approach is likely safe for an

286

uncomplicated TKA with the appropriate health status. The findings of this study have resulted

287

in the authors continued use of a more restricted, patient driven perioperative fluid protocol for

288

appropriate patients undergoing total joint arthroplasty.

289

pg. 14

290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335

REFERENCES 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

Holte K, Klarskov B, Christensen DS, Lund C, Nielsen KG, Bie P, et al. Liberal versus restrictive fluid administration to improve recovery after laparoscopic cholecystectomy: A randomized, double-blind study. Ann Surg. 2004;240(5):892–9. Bundgaard-Nielsen M, Secher NH, Kehlet H. “Liberal” vs. “restrictive” perioperative fluid therapy - A critical assessment of the evidence: Review Article. Acta Anaesthesiol Scand. 2009;53(7):843–51. Corcoran T, Emma Joy Rhodes J, Clarke S, Myles PS, Ho KM. Perioperative fluid management strategies in major surgery: A stratified meta-analysis. Anesth Analg. 2012;114(3):640–51. Myles PS, Bellomo R, Corcoran T, Forbes A, Peyton P, Story D, et al. Restrictive versus Liberal Fluid Therapy for Major Abdominal Surgery. N Engl J Med. 2018;378(24):2263– 74. Omesiete NP, Martinez C, Pandit V, Villalvazo Y, Thompson EL, Nocera M, et al. Restricting Intraoperative Fluid Volume Allows Earlier Return of Bowel Function after Colon Surgery: A Prospective Matched-Cohort Analysis. J Am Coll Surg [Internet]. 2018;227(4):S74–5. Available from: https://doi.org/10.1016/j.jss.2019.06.013 Zhang X, Zheng W, Chen C, Kang X, Zheng Y, Bao F, et al. Goal-directed fluid therapy does not reduce postoperative ileus in gastrointestinal surgery A meta-analysis of randomized controlled trials. Med (United States). 2018;97(45). Srinivasa S, Taylor MHG, Singh PP, Yu TC, Soop M, Hill AG. Randomized clinical trial of goal-directed fluid therapy within an enhanced recovery protocol for elective colectomy. Br J Surg. 2013;100(1):66–74. Rollins KE, Lobo DN. Intraoperative goal-directed fluid therapy in elective major abdominal surgery : A meta-analysis of randomized controlled trials. Ann Surg. 2016;263(3):465–76. Xu C, Peng J, Liu S, Huang YZ, Guo XW, Xiao HB, et al. Goal-directed fluid therapy versus conventional fluid therapy in colorectal surgery: A meta analysis of randomized controlled trials. Int J Surg [Internet]. 2018;56(April 2018):264–73. Available from: https://doi.org/10.1016/j.ijsu.2018.06.034 Holte K, Kristensen BB, Valentiner L, Foss NB, Husted H, Kehlet H. Liberal versus restrictive fluid management in knee arthroplasty: A randomized, double-blind study. Anesth Analg. 2007;105(2):465–74. Makaryus R, Miller TE, Gan TJ. Current concepts of fluid management in enhanced recovery pathways. Br J Anaesth [Internet]. 2018;120(2):376–83. Available from: https://doi.org/10.1016/j.bja.2017.10.011 Schol PBB, Terink IM, Lancé MD, Scheepers HCJ. Liberal or restrictive fluid management during elective surgery: a systematic review and meta-analysis. J Clin Anesth [Internet]. 2016 Dec [cited 2019 Jul 26];35:26–39. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0952818016303725 Sanford EL, Zurakowski D, Litvinova A, Zalieckas JM, Cravero JP. The association between high‐volume intraoperative fluid administration and outcomes among pediatric patients undergoing large bowel resection. Anderson B, editor. Pediatr Anesth [Internet]. 2019 Apr 4 [cited 2019 Jul 26];29(4):315–21. Available from: https://onlinelibrary.wiley.com/doi/abs/10.1111/pan.13581 pg. 15

336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357

14.

15. 16. 17.

18. 19.

20.

Cecconi M, Fasano N, Langiano N, Divella M, Costa MG, Rhodes A, et al. Goal-directed haemodynamic therapy during elective total hip arthroplasty under regional anaesthesia. Crit Care [Internet]. 2011 [cited 2019 Jul 26];15(3):R132. Available from: http://ccforum.biomedcentral.com/articles/10.1186/cc10246 Rocca G Della, Pompei L. Goal-directed therapy in anesthesia: Any clinical impact or just a fashion? Minerva Anestesiol. 2011;77(5):545–53. Pichonnaz C, Bassin JP, Currat D, Martin E, Jolles BM. Bioimpedance for oedema evaluation after total knee arthroplasty. Physiother Res Int. 2013;18:140–7. Pichonnaz C, Bassin J-P, Lécureux E, Currat D, Jolles BM. Bioimpedance spectroscopy for swelling evaluation following total knee arthroplasty: a validation study. BMC Musculoskelet Disord. 2015;16(1):100–8. Pua YH. The Time Course of Knee Swelling Post Total Knee Arthroplasty and Its Associations with Quadriceps Strength and Gait Speed. J Arthroplasty. 2015;30:1215–9. Halawi MJ, Caminiti N, Cote MP, Lindsay AD, Williams VJ. The Most Significant Risk Factors for Urinary Retention in Fast-track Total Joint Arthroplasty are Iatrogenic. J Arthroplasty [Internet]. 2019 Jan [cited 2019 Jul 26];34(1):136–9. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0883540318307484 Lawrie CM, Ong AC, Hernandez VH, Rosas S, Post ZD, Orozco FR. Incidence and Risk Factors for Postoperative Urinary Retention in Total Hip Arthroplasty Performed Under Spinal Anesthesia. J Arthroplasty [Internet]. 2017 Dec [cited 2019 Aug 22];32(12):3748– 51. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0883540317305922

pg. 16

358 359 360

Figure Legend Figure 1. Patient flow diagram.

pg. 17

Table 1. Patient demographics and intraoperative findings Patient Demographics Traditional “IV” Fluid Group Age 61.2 ± 7.5 Gender Male – 41.5 % Female – 58.5 % BMI 28.5 ± 4.9 Procedure Time 74.9 ± 15.1 Blood Loss 136.2 ± 46.1

Oral Hydration Group 60.9 ± 6.8 Male – 44.6 % Female – 55.4 % 28.1 ± 4.2 78.9 ± 17.9 128.5 ± 47.4

p value p = 0.826 p = 0.723 P = 0.581 p = 0.164 p = 0.350

Table 2. Pre-operative and post-operative urine specific gravity Urine Specific Gravity at Pre- OP <1.005 1.010 1.015 1.020 1.025 >1.030 Urine Specific Gravity at POD 1 <1.005 1.010 1.015 1.020 1.025 >1.030 Urine Specific Gravity at 2 Week F/U <1.005 1.010 1.015 1.020 1.025 >1.030

Traditional “IV” Fluid Group (%) 5 (7.7) 14 (21.5) 15 (23.1) 12 (18.5) 14 (21.5) 5 (7.7)

Oral Hydration Group (%) 21 (32.8) 16 (25.0) 14 (21.9) 5 (7.8) 5 (7.8) 3 (4.7)

p value

17 (27.0) 13 (20.6) 13 (20.6) 6 (9.5) 7 (11.1) 7 (11.1)

20 (31.3) 11 (17.2) 8 (12.5) 4 (6.3) 11 (17.2) 10 (15.6)

p = 0.633

12 (18.5) 10 (15.4) 6 (9.2) 9 (13.8) 11 (16.9) 17 (26.2)

13 (20.6) 7 (11.1) 11 (17.5) 9 (14.3) 9 (14.3) 14 (22.2)

p = 0.773

p = 0.002

Table 3. Pre-operative vs. post-operative differences in body weight Change in Body Weight Change from Pre- Op to

Traditional “IV” Fluid Group Mean ± SD 7.0 ± 4.3

Oral Hydration p value Group Mean ± SD 3.0 ± 3.9 p<0.0001 pg. 18

Evening of Surgery Change from Pre- Op to Post 9.1 ± 4.3 Op Day 1 Change from Pre- Op to 48 6.2 ± 5.0 after Discharge Change from Pre- Op to end 1.7 ± 4.2 of Week 1 Change from Pre- Op to end -1.9 ± 4.4 of Week 2

4.7 ± 3.9

p<0.0001

4.4 ± 4.0

p = 0.032

1.3 ± 4.2

p = 0.605

-1.6 ± 4.4

P = 0.739

Table 4. Bioelectrical impedance measurements Bioelectrical Impedance change from Pre- Op Post Op Day 1 2 Weeks Post Op 6 Weeks Post Op

Traditional “IV” Fluid Group Mean ± SD -17.8 ± 30.3 -41.4 ± 34.2 -30.9 ± 25.6

Oral Hydration p value Group Mean ± SD 4.2 ± 29.7 p<0.0001 -38.5 ± 45.7 p = 0.686 -31.0 ± 31.4 p = 0.988

Table 5. Lower extremity girth measurements Change in leg girth

Traditional “IV” Fluid Group Mean ± SD Change in girth 10 cm Superior from Pre- Op Post Op Day 1 2.5 ± 3.0 2 Weeks Post Op 0.9 ± 3.0 6 Weeks Post Op 0.0 ± 2.6 Change in girth 5 cm Superior from Pre- Op Post Op Day 1 4.1 ± 2.4 2 Weeks Post Op 2.0 ± 2.3 6 Weeks Post Op 0.9 ± 1.9 Change in girth mid patella from Pre- Op Post Op Day 1 4.6 ± 2.2 2 Weeks Post Op 2.2 ± 2.3 6 Weeks Post Op 1.3 ± 1.6 Change in girth 10 cm inferior patella from Pre- Op Post Op Day 1 2.7 ± 2.7 2 Weeks Post Op 0.8 ± 2.1 6 Weeks Post Op -0.2 ± 1.3

Oral Hydration Group Mean ± SD

p value

2.1 ± 3.1 1.6 ± 4.0 -0.0 ± 3.3

p = 0.383 p = 0.259 p = 0.889

4.0 ± 2.6 2.8 ± 2.6 1.0 ± 2.2

p = 0.807 p = 0.083 p = 0.896

4.1 ± 2.6 3.0 ± 2.0 1.3 ± 2.0

p = 0.278 p = 0.050 p = 0.922

1.5 ± 4.6 0.9 ± 2.3 0.1 ± 1.5

p = 0.076 p = 0.765 p = 0.332

Table 6. VR-12, KOOS, KS scores

pg. 19

PRO’s Change from Pre-Op Traditional “IV” Fluid Group Mean ± SD VR-12 Change from Pre- Op 2 Weeks Post Op Mental Score -4.7 ± 12.2 2 Weeks Post Op Physical -8.3 ± 10.0 Score 6 Weeks Post Op Mental Score -2.5 ± 11.5 6 Weeks Post Op Physical -0.5 ± 10.6 Score KSS change from Pre- Op 2 Weeks Post Op Pain Score 8.1 ± 17.4 2 Weeks Post Op Functional -29.9 ± 26.3 Score 2 Weeks Post Op Total Score -14.6 ± 39.9 6 Weeks Post Op Pain Score 16.4 ± 19.0 6 Weeks Post Op Functional 1.8 ± 23.8 Score 6 Weeks Post Op Total Score 31.9 ± 36.3 KOOS JR Change from Pre- Op 2 Weeks Post Op -1.6 ± 16.2 6 Weeks Post Op 9.4 ± 14.5

Oral Hydration Group Mean ± SD

p value

-1.2 ± 10.4 -4.5 ± 11.2

p = 0.087 p = 0.046

-3.2 ± 11.4 2.2 ± 12.2

p = 0.738 p = 0.195

12.5 ± 16.3 -21.3 ± 24.9

p = 0.145 p = 0.060

-0.2 ± 36.2 18.5 ± 15.3 7.1 ± 24.0

p = 0.034 p = 0.485 p = 0.212

40.5 ± 35.4

p = 0.175

3.1 ± 13.8 12.3 ± 14.1

p = 0.078 p = 0.263

Table 7. Blood pressure measurements Change in Blood Pressure Pre Op to Recovery Room (Systolic) Pre Op to Recovery Room (Diastolic) Pre Op to Arrival on 4th floor (Systolic) Pre Op to Arrival on 4th floor (Diastolic) Pre Op to 08:00 on floor (Systolic) Pre Op to 08:00 on floor (Diastolic) Pre Op to 16:00 on floor (Systolic) Pre Op to 16:00 on floor (Diastolic) Pre Op to 24:00 on floor (Systolic)

Traditional “IV” Fluid Group Mean ± SD 21.5 ± 15.9

Oral Hydration p value Group Mean ± SD 26.0 ± 16.7 p = 0.114

18.4 ± 12.2

17.2 ± 11.9

p = 0.574

10.6 ± 12.8

19.4 ± 13.5

p<0.0001

10.5 ± 10.3

9.3 ± 10.8

p = 0.526

17.0 ± 12.9

23.4 ± 13.3

p = 0.006

16.5 ± 9.3

16.0 ± 9.6

p = 0.772

20.9 ± 14.0

25.8 ± 13.8

p = 0.046

16.3 ± 11.2

18.7 ± 11.8

p = 0.235

12.5 ± 14.7

12.5 ± 12.9

p = 0.986

pg. 20

Pre Op to 24:00 on floor (Diastolic)

14.6 ± 9.9

11.1 ± 10.2

p = 0.197

Table 8. Laboratory values Change in Labs from Pre-Op Traditional “IV” Fluid Group Mean ± SD Change from Pre- Operative Labs to POD 1 Hematocrit 7.9 ± 2.6 BUN 2.6 ± 4.4 Calcium 0.9 ± 0.4 Carbon Dioxide 0.4 ± 2.8 Chloride 0.6 ± 13.4 Creatinine 0.0 ± 0.1 Glucose -29.6 ± 21.7 Potassium -0.2 ± 0.4 Sodium -0.3 ± 2.3 Change from Pre- Operative Labs to 2 Week Visit Hematocrit 6.1 ± 7.5 BUN -1.6 ± 6.2 Calcium 0.3 ± 1.6 Carbon Dioxide 0.2 ± 5.6 Chloride 5.1 ± 19.1 Creatinine -0.0 ± 0.2 Glucose 2.5± 20.4 Potassium 0.1 ± 0.9 Sodium 5.9 ± 24.9

Oral Hydration Group Mean ± SD

p value

7.3 ± 2.8 0.5 ± 4.7 0.6 ± 0.4 0.8 ± 2.4 2.5 ± 3.1 -0.1 ± 0.1 -35.9 ± 29.7 -0.3 ± 0.5 2.5 ± 2.9

p = 0.240 p = 0.011 p<0.0001 p = 0.352 p = 0.265 p<0.0001 p = 0.166 p = 0.399 p<0.0001

6.5 ± 5.8 -1.2 ± 4.9 0.0 ± 0.4 -0.1 ± 2.6 1.2 ± 2.5 -0.0 ± 0.1 -3.9 ± 16.4 -0.1 ± 0.4 0.9 ± 2.1

p = 0.695 p = 0.715 p = 0.263 p = 0.672 p = 0.112 p = 0.463 p = 0.055 p = 0.159 p = 0.121

pg. 21