Perioperative Differences Between Outpatient and Inpatient Pathways Following Hip and Knee Arthroplasty

Abstract

Background

Optimization of clinical pathways and logistics led to the introduction of outpatient joint arthroplasty of the hip and knee. Nevertheless, little is known about what these current protocols look like and how they differ from "standard" inpatient protocols. This study aimed to find preoperative, intraoperative, and postoperative differences between outpatient and inpatient pathways.

Methods

A questionnaire (ranging between 23 and 37 items) was developed and administered by email to orthopedic surgeons who were a member of the Dutch Hip Society and Dutch Knee Society. Survey response rate was 38% (N = 117).

Results

No significant differences were found in preoperative pathway characteristics. The administration regime for tranexamic acid significantly differed between outpatient and inpatient pathways (P < .001 and P = .002 for hip and knee arthroplasty, respectively), with outpatient pathways using a combined (eg, oral and intravenous) administration regime more frequently. The perioperative antibiotic prophylaxis regime also significantly differed between outpatient and inpatient pathways (P < .001 and P = .014, respectively), with outpatient pathways more frequently incorporating fewer antibiotic doses. Same-day postoperative mobilization significantly less often occurred if surgery took place later that day in inpatient hip arthroplasty pathways (24%; P = .034). Postoperative hemoglobin-check occurred significantly more often on indication in outpatient than in inpatient hip and knee arthroplasty pathways (∼75% vs ∼25%; P = .001).

Conclusions

Few intraoperative and postoperative differences in outpatient and inpatient pathways were found and probably mainly relied on logistical grounds. Nonetheless, findings suggested that outpatient pathways tended to be more up-to-date and innovative than inpatient pathways.

Introduction

Clinical pathways were introduced in the United States in the 1980s to cut the costs and to reduce variation in care that does not influence outcome. It is basically an algorithm that outlines a structured evidence-based sequence and timing of interventions for the care of a specific group of patients with a particular diagnosis or procedure. The purpose is to standardize procedures to ensure an optimal and reproducible outcome regarding quality, efficiency, and economic saving. Nowadays these pathways are widely implemented in high-volume, high-cost, and high-predictable care processes [1].

A variant on a clinical pathway was developed by Kehlet in an effort to improve recovery after colorectal surgery: enhanced recovery after surgery (ERAS) [2]. He hypothesized that the surgical stress response was the key pathogenic factor in postoperative morbidity and contributed to “organ dysfunction” by metabolic and endocrine derangements (eg, pain, postoperative nausea and vomiting, fatigue). As no single surgical/anesthetic technique or drug regimen had been shown to eliminate postoperative morbidity, the ERAS concept promoted a multimodal approach to address the surgical stress response with multidisciplinary team effort [3]. In the last 2 decades, these “fast-track” pathways have also been implemented in hip and knee arthroplasty and successfully proved to enhance recovery and patient satisfaction with care, decrease morbidity, and effectively reduce hospital length of stay (LOS) and healthcare costs [4,5].

Auditing is another important feature of the “fast-track” concept, constantly revisiting and evaluating routines and current practices in an attempt to incorporate new evidence-based clinical features in the perioperative pathway to promote recovery. As a result of these continuous improvements in perioperative care—in combination with optimized pathway and hospital logistics—outpatient joint arthroplasty (OJA; discharging patients on the same day of surgery) pathways emerged in several clinics [6,7]. These OJA pathways have been proven to be feasible and safe in a selected group of patients, and so they are gaining popularity [[8], [9], [10]].

Although evidence and consensus statements exist for the organization of perioperative care in total joint arthroplasty (TJA) of the hip and knee, there are no studies comparing perioperative features of outpatient vs standard inpatient pathways. The primary research purpose of this study was to assess preoperative, intraoperative, and postoperative differences between outpatient and standard inpatient pathways following hip and knee arthroplasty in The Netherlands. We hypothesized that outpatient pathways contained more state-of-the-art features of perioperative care when compared to inpatient pathways.

Material and methods

This study was approved by the local institutional review board (study number Z20210053).

Setting and subjects

This study was conducted as a follow-up survey among the same study population (active members of the Dutch Hip Society and the Dutch Knee Society) as described previously by our research team [11].

Questionnaire

A new questionnaire was developed and concerns a follow-up to a previous questionnaire (on the surgeon’s perspective on the implementation of OJA), the results of which have already been published [11]. The questionnaire comprised 4 sections to determine surgeon’s baseline characteristics and to identify preoperative, intraoperative, and postoperative features of current outpatient or inpatient pathways. If the respondent performed OJA, then all subsequent questions were intended to specify their outpatient pathway (and vice versa). Outpatient was defined as patients who were discharged to their own home environment on the day of surgery and who do not have an overnight stay at either the hospital or another non-home facility [9,11,12].

The number of questionnaire items varied individually between 23 and 37 based on the responses of the participants. An English version of the questionnaire is presented in Appendix A.

Statistics

All analyses were performed with SPSS (IBM Corp. Released 2019. IBM SPSS Statistics for Windows, Version 26.0; IBM Corp., Armonk, NY).

Descriptives were displayed as means with standard deviations (and range) for continuous variables and frequencies with percentages for categorical variables.

Differences between outpatient and inpatient pathways were assessed using chi-square or (in case of cells with an expected value < 5) Fisher’s exact test for categorical variables and an unpaired t-test for continuous variables. A two-sided P-value ≤.05 was considered to indicate statistical significance.

Study population

Our prior administered survey was completed by 123 of 331 orthopedic surgeons [11]. One hundred seventeen of 123 respondents (95%) completed the follow-up survey and were included in the analysis. This resulted in a response rate of 35.3% for the entire study population. Demographics of respondents are summarized in Table 1.

Table 1.

Demographics of the respondents.

Characteristic	Total	OJA	Non-OJA	P value
Number of respondents (N)	117	20	97
Age (y)	46 ± 8.9 (32-66)	47 ± 9.3 (32-66)	46 ± 8.8 (33-63)	.74
Sex				1.0
Female	3 (2.6%)	0 (0%)	3 (3.1%)
Male	114 (97%)	20 (100%)	94 (97%)
Affiliation				.13
Academic (university) hospital	13 (11%)	0 (0%)	13 (13%)
Non-academic teaching hospital	54 (46%)	14 (70%)	40 (41%)
Non-academic non-teaching hospital	33 (28%)	4 (20%)	29 (30%)
Private clinic	13 (11%)	2 (10%)	11 (11%)
Other	4 (3.4%)	0 (0%)	4 (4.1%)
Orthopedic (sub) specialty				.25
Hip and knee arthroplasty	68 (58%)	13 (65%)	55 (57%)
Only hip arthroplasty	15 (13%)	4 (20%)	11 (11%)
Only knee arthroplasty	34 (29%)	3 (15%)	31 (32%)
Hip arthroplasty groupa
N, surgeons	83	17	66
N, surgeries (per year)	116 ± 62 (20-330)	150 ± 71 (70-330)	108 ± 57 (20-280)	.011
Types of hip arthroplasty
Primary	83 (100%)	17 (100%)	66 (100%)	N/a
Revision	60 (72%)	13 (76%)	47 (71%)	.77
Knee arthroplasty groupa
N, surgeons	102	16	86	.42
N, surgeries (per year)	102 ± 66 (10-380)	114 ± 46 (40-180)	100 ± 69 (10-380)
Types of knee arthroplasty
Primary	102 (100%)	16 (100%)	86 (100%)	N/a
Revision	64 (63%)	10 (63%)	54 (63%)	.98

OJA, outpatient joint arthroplasty.

Discrete variables are given as frequency (percentage), and continuous variables as mean ± standard deviation (range); bold values indicate statistically significant difference (P < .05).

^aRespondents that perform “hip AND knee arthroplasty” are included in both the hip arthroplasty group and the knee arthroplasty group.

Results

Preoperative pathway characteristics

Informative leaflets were the most frequently used type of education tool. Prehabilitation was used by ∼35% of respondents in both hip and knee arthroplasty procedures. No major differences were found between outpatient and inpatient pathways (Table 2, Table 3).

Table 2.

Preoperative characteristics of the outpatient and inpatient hip arthroplasty pathway.

Characteristic	Total	Outpatient	Inpatient	P value
Number of respondents	83	17	66
Preoperative education	83 (100%)	17 (100%)	66 (100%)	N/a
Type of education useda				N/a
Verbal	61 (73%)	14 (82%)	47 (71%)
Informative leaflets	78 (94%)	17 (100%)	61 (92%)
Videos	35 (42%)	5 (29%)	30 (45%)
Decision aids	33 (40%)	7 (41%)	26 (39%)
eHealth (eg, patient journey app)	24 (29%)	4 (24%)	20 (30%)
Other	5 (6.0%)	1 (5.9%)	4 (6.1%)
Pre-habilitation	27 (33%)	8 (47%)	19 (29%)	.16
Preoperative screening	66 (80%)	15 (88%)	51 (77%)	.50
Type of screening tools useda				N/a
Charlson Comorbidity Index (CCI)	3 (4.5%)	0 (0%)	3 (5.9%)
American Society of Anesthesiologists (ASA) class	57 (86%)	12 (80%)	45 (88%)
Outpatient Arthroplasty Risk Assessment (OARA)	1 (1.5%)	0 (0%)	1 (2.0%)
Total joint replacement (TJR) risk calculator	1 (1.5%)	0 (0%)	1 (2.0%)
Readmission Risk Assessment Tool (RRAT)	1 (1.5%)	0 (0%)	1 (2.0%)
Protocollary defined selection criteria	14 (21%)	7 (47%)	7 (14%)
Other	7 (11%)	3 (20%)	4 (7.8%)

Discrete variables are given as frequency (percentage).

^aMultiple answers were allowed. Proportions of respondents that made use of this feature were displayed.

Table 3.

Preoperative characteristics of the outpatient and inpatient knee arthroplasty pathway.

Characteristic	Total	Outpatient	Inpatient	P value
Number of respondents	102	16	86
Preoperative education	102 (100%)	16 (100%)	86 (100%)	N/a
Type of education useda				N/a
Verbal	77 (75%)	14 (88%)	63 (73%)
Informative leaflets	96 (94%)	16 (100%)	80 (93%)
Videos	43 (42%)	3 (19%)	40 (47%)
Decision aids	39 (38%)	6 (38%)	33 (38%)
eHealth (eg, patient journey app)	28 (27%)	4 (25%)	24 (28%)
Other	5 (4.9%)	1 (6.3%)	4 (4.7%)
Pre-habilitation	37 (36%)	9 (56%)	28 (33%)	.091
Preoperative screening	81 (79%)	14 (88%)	67 (78%)	.51
Type of screening tools useda				N/a
Charlson Comorbidity Index (CCI)	4 (4.9%)	0 (0%)	4 (6.0%)
American Society of Anesthesiologists (ASA) class	73 (90%)	12 (86%)	60 (90%)
Outpatient Arthroplasty Risk Assessment (OARA)	1 (1.2%)	0 (0%)	1 (1.5%)
Total joint replacement (TJR) risk calculator	2 (2.5%)	0 (0%)	2 (3.0%)
Readmission Risk Assessment Tool (RRAT)	1 (1.2%)	0 (0%)	1 (1.5%)
Protocollary defined selection criteria	22 (27%)	6 (43%)	16 (24%)
Other	8 (9.9%)	3 (21%)	5 (7.5%)

Discrete variables are given as frequency (percentage).

^aMultiple answers were allowed. Proportions of respondents that made use of this feature were displayed.

Approximately 80% of respondents made use of preoperative screening tools. The most used screening tool was the American Society of Anesthesiologists (ASA) classification. Protocollary defined selection criteria were more frequently used in outpatient (∼45%) compared with inpatient hip and knee arthroplasty (14% and 24%, respectively) pathways (Table 2, Table 3).

Intraoperative pathway characteristics

Anesthesia

Spinal anesthesia (with or without sedation) was the most frequently used type of anesthesia (∼75%). Little use of peripheral nerve (eg, femoral/saphenous) blocks (PNBs) was reported and was only limited to inpatient pathways (6% in hip arthroplasty and 24% in knee arthroplasty). The use of local infiltration anesthesia (LIA) was common in knee arthroplasty (96%), but not in hip arthroplasty pathways (7.2%). No differences were found between outpatient and inpatient pathways (Table 4, Table 5).

Table 4.

Intraoperative characteristics of the outpatient and inpatient hip arthroplasty pathway.

Characteristic	Total	Outpatient	Inpatient	P value
Number of respondents	83	17	66
Anesthesia, preference				.55
General	13 (16%)	2 (12%)	11 (17%)
Spinal with sedation	28 (34%)	6 (35%)	22 (33%)
Spinal	34 (41%)	9 (53%)	25 (38%)
No preference	7 (8.4%)	0 (0%)	7 (11%)
Other	1 (1.2%)	0 (0%)	1 (1.5%)
Surgical approach, preference				.057
Posterolateral	48 (58%)	6 (35%)	42 (64%)
Direct anterior	29 (35%)	9 (53%)	19 (29%)
Straight lateral	3 (3.6%)	0 (0%)	3 (4.5%)
Anterolateral	1 (1.2%)	1 (5.9%)	0 (0%)
Direct superior	3 (3.6%)	1 (5.9%)	2 (3.0%)
Peripheral nerve blocks				.8
Lumbar plexus block	1 (1.2%)	0 (0%)	1 (1.5%)
Fascia iliaca block	1 (1.2%)	0 (0%)	1 (1.5%)
Other	2 (2.4%)	0 (0%)	2 (3.0%)
No	79 (95%)	17 (100%)	62 (94%)
LIA use	6 (7.2%)	1 (5.9%)	5 (7.6%)	1.0
Composition of LIA useda				N/a
Ropivacaine	5 (83%)	1 (100%)	4 (80%)
Bupivacaine	1 (17%)	0 (0%)	1 (20%)
Adrenaline	2 (33%)	0 (0%)	2 (40%)
Tranexamic acid use	83 (100%)	17 (100%)	66 (100%)	N/a
When				.091
Preoperative	28 (34%)	3 (17%)	25 (38%)
Intraoperative	9 (11%)	0 (0%)	9 (14%)
Postoperative	3 (3.6%)	0 (0%)	3 (4.5%)
Preoperative and intraoperative	14 (17%)	3 (17%)	11 (17%)
Preoperative and postoperative	21 (25%)	8 (47%)	13 (20%)
Intraoperative and postoperative	4 (4.8%)	2 (12%)	2 (3.0%)
Preoperative, intraoperative and postoperative	4 (4.8%)	1 (5.9%)	3 (4.5%)
How				<.001
Oral	3 (3.6%)	2 (12%)	1 (1.5%)
Intravenous	61 (73%)	6 (35%)	55 (83%)
Intra-articular	0 (0%)	0 (0%)	0 (0%)
Oral and intravenous	11 (13%)	8 (47%)	3 (4.5%)
Intravenous and intra-articular	7 (8.4%)	1 (5.9%)	6 (9.1%)
I don't know	1 (1.2%)	0 (0%)	1 (1.5%)
Drain use	1 (1.2%)	0 (0%)	1 (1.5%)	1.0
Urinary catheter use	2 (2.4%)	0 (0%)	2 (3.0%)	1.0
Antibiotics				<.001
1 dose	13 (16%)	2 (12%)	11 (17%)
2 doses	11 (13%)	9 (53%)	2 (3.0%)
3 doses	58 (70%)	6 (35%)	52 (79%)
Other	1 (1.2%)	0 (0%)	1 (1.5%)

LIA, local infiltration anesthesia.

Discrete variables are given as frequency (percentage); bold values indicate statistically significant difference (P < .05).

^aMultiple answers were allowed. Proportions of respondents that made use of this feature were displayed.

Table 5.

Intraoperative characteristics of the outpatient and inpatient knee arthroplasty pathway.

Characteristic	Total	Outpatient	Inpatient	P value
Number of respondents	102	16	86
Anesthesia, preference				.43
General	5 (4.9%)	0 (0%)	5 (5.8%)
Spinal with sedation	36 (35%)	7 (44%)	29 (34%)
Spinal	40 (39%)	8 (50%)	32 (37%)
No preference	18 (18%)	1 (6.3%)	17 (20%)
Other	3 (2.9%)	0 (0%)	3 (3.5%)
Surgical approach, preference				.061
Medial parapatellar	100 (98%)	15 (94%)	85 (99%)
Midvastus	1 (0.98%)	1 (6.3%)	0 (0%)
Subvastus	1 (0.98%)	0 (0%)	1 (1.2%)
Tourniquet use	46 (45%)	6 (38%)	40 (47%)	.51
Technique				.22
During the entire operation	32 (70%)	6 (100%)	26 (65%)
During cementation	11 (24%)	0 (0%)	11 (28%)
Other	3 (6.5%)	0 (0%)	3 (7.5%)
Peripheral nerve blocks				.30
Femoral nerve block	8 (7.8%)	0 (0%)	8 (9.3%)
Saphenous nerve (adductor canal) block	8 (7.8%)	0 (0%)	8 (9.3%)
Obturator nerve block	1 (0.98%)	0 (0%)	1 (1.2%)
Other	4 (3.9%)	0 (0%)	4 (4.7%)
No	81 (79%)	16 (100%)	65 (76%)
LIA use	98 (96%)	16 (100%)	82 (95%)	1.0
Composition of LIA useda				N/a
Ropivacaine	85 (87%)	15 (94%)	70 (85%)
Bupivacaine	13 (13%)	1 (6.3%)	12 (15%)
Ketorolac	4 (4.1%)	1 (6.3%)	3 (3.7%)
Adrenaline	31 (32%)	3 (19%)	28 (34%)
Morphine	1 (1.0%)	0 (0%)	1 (1.2%)
Corticosteroids	3 (3.1%)	1 (6.3%)	2 (2.4%)
Don't know	2 (2.0%)	0 (0%)	2 (2.4%)
Other	1 (1.0%)	0 (0%)	1 (1.2%)
Tranexamic acid use	102 (100%)	16 (100%)	86 (100%)	N/a
When				.052
Preoperative	35 (34%)	2 (13%)	33 (38%)
Intraoperative	12 (12%)	0 (0%)	12 (14%)
Postoperative	2 (2.0%)	0 (0%)	2 (2.3%)
Preoperative and intraoperative	16 (16%)	3 (19%)	13 (15%)
Preoperative and postoperative	28 (27%)	8 (50%)	20 (23%)
Intraoperative and postoperative	4 (3.9%)	2 (13%)	2 (2.3%)
Preoperative, intraoperative and postoperative	5 (4.9%)	1 (6.3%)	4 (4.7%)
How				.002
Oral	6 (5.9%)	1 (6.3%)	5 (5.8%)
Intravenous	72 (71%)	6 (38%)	66 (77%)
Intra-articular	0 (0%)	0 (0%)	0 (0%)
Oral and intravenous	15 (15%)	8 (50%)	7 (8.1%)
Intravenous and intra-articular	7 (6.9%)	1 (6.3%)	6 (7.0%)
Oral, intravenous and intra-articular	1 (0.98%)	0 (0%)	1 (1.2%)
I don't know	1 (0.98%)	0 (0%)	1 (1.2%)
Drain use	2 (2.0%)	0 (0%)	2 (2.3%)	1.0
Urinary catheter use	2 (2.0%)	0 (0%)	2 (2.3%)	1.0
Antibiotics				.014
1 dose	17 (17%)	1 (6.3%)	16 (19%)
2 doses	17 (17%)	7 (44%)	10 (12%)
3 doses	66 (65%)	8 (50%)	58 (67%)
Other	2 (2.0%)	0 (0%)	2 (2.3%)

LIA, local infiltration anesthesia.

Discrete variables are given as frequency (percentage); bold values indicate statistically significant difference (P < .05).

^aMultiple answers were allowed. Proportions of respondents that made use of this feature were displayed.

Surgical techniques

The posterolateral (PL) approach was preferred by most respondents (58%) in total hip arthroplasty (THA). Higher rates of the direct anterior approach (DAA) were observed in outpatient than in inpatient hip arthroplasty pathways (53% vs 29%, P = .057). Most respondents (98%) preferred the medial parapatellar (MMP) approach in total knee arthroplasty (TKA; Table 4, Table 5).

Intraoperative medication

Tranexamic acid (TXA) was used by all respondents. Great variability—in when (perioperative timing) and how (route) TXA was administered—was observed in both hip and knee arthroplasty pathways (Table 4, Table 5). Significant differences between outpatient and inpatient pathways on route of administration were found in both hip and knee arthroplasty (P < .001 and P = .002, respectively).

Also, great variability was observed in the use of antibiotic prophylaxis (Table 4, Table 5). Significant differences in the number of antibiotic doses were found between outpatient and inpatient pathways in both hip and knee arthroplasty (P < .001 and P = .014, respectively).

Postoperative pathway characteristics

The timing of postoperative mobilization significantly differed between outpatient and inpatient pathways for hip arthroplasty (P = .034). In 24% of inpatient pathways, mobilization was delayed to the next day if surgery took place later that day (Table 6).

Table 6.

Postoperative characteristics of the outpatient and inpatient hip arthroplasty pathway.

Characteristic	Total	Outpatient	Inpatient	P value
Number of respondents	83	17	66
Average length of stay after standard inpatient THA (days)a	2.1 ± 0.80 (1-6)	2.1 ± 0.66 (1-3)	2.1 ± 0.84 (1-6)	.73
Anticoagulant use	83 (100%)	17 (100%)	66 (100%)	N/a
Preferencesb				N/a
Low-molecular-weight heparins (LMWHs)	66 (80%)	14 (82%)	52 (79%)
Novel oral anticoagulants (NOACs)/Direct oral anticoagulants (DOACs)	24 (29%)	4 (24%)	20 (30%)
Platelet aggregation inhibitor	2 (2.4%)	0 (0%)	2 (3.0%)
Length of use				.12
1-3 weeks	2 (2.4%)	1 (5.9%)	1 (1.5%)
4-6 weeks	77 (93%)	15 (88%)	62 (94%)
>6 weeks	1 (1.2%)	1 (5.9%)	0 (0%)
Other	3 (3.6%)	0 (0%)	3 (4.5%)
Morphine use				.42
Rescue medication	48 (58%)	12 (71%)	36 (55%)
Standard/scheduled medication	26 (31%)	4 (24%)	22 (33%)
Patient controlled analgesia	5 (6.0%)	0 (0%)	5 (7.6%)
No	2 (2.4%)	0 (0%)	2 (3.0%)
Other	2 (2.4%)	1 (5.9%)	1 (1.5%)
Physical therapy				.034
On the same day	67 (81%)	17 (100%)	50 (76%)
Not the same day if surgery takes place later on the day	16 (19%)	0 (0%)	16 (24%)
Hb check				<.001
Always	54 (65%)	4 (24%)	50 (76%)
Only if indicated	28 (34%)	12 (71%)	16 (24%)
No	1 (1.2%)	1 (5.9%)	0 (0%)
X-ray				.19
Direct postoperative	77 (93%)	15 (88%)	62 (94%)
During follow-up visit	2 (2.4%)	0 (0%)	2 (3.0%)
Only if indicated	1 (1.2%)	0 (0%)	1 (1.5%)
Other	3 (3.6%)	2 (12%)	1 (1.5%)

Discrete variables are given as frequency (percentage), and continuous variables as mean ± standard deviation (range); bold values indicate statistically significant difference (P < .05).

^aBoth groups of orthopedic surgeons (performing both inpatient and outpatient joint arthroplasty vs only inpatient joint arthroplasty) estimated their average LOS for the standard inpatient total joint pathways.

^bMultiple answers were allowed. Proportions of respondents that made use of this feature were displayed.

Postoperative medication

The majority of respondents preferred low-molecular-weight heparins as a type of anticoagulant (∼80%) and with a 4- to 6-week length of anticoagulant use (∼93%). The preferred length of anticoagulant prophylaxis significantly differed between outpatient and inpatient pathways for knee arthroplasty (P = .043), but not for hip arthroplasty (P = .12) (Table 6, Table 7).

Table 7.

Postoperative characteristics of the outpatient and inpatient knee arthroplasty pathway.

Characteristic	Total	Outpatient	Inpatient	P value
Number of respondents	102	16	86
Average length of stay after standard inpatient TKA (days)a	2.3 ± 0.85 (1-6)	2.1 ± 0.62 (1-3)	2.3 ± 0.89 (1-6)	.51
Anticoagulant use	102 (100%)	16 (100%)	86 (100%)	N/a
Preferencesb				N/a
Low-molecular-weight heparins (LMWHs)	83 (81%)	14 (81%)	69 (80%)
Novel oral anticoagulants (NOACs)/Direct oral anticoagulants (DOACs)	27 (26%)	2 (13%)	25 (29%)
Platelet aggregation inhibitor	6 (5.9%)	1 (6.3%)	5 (5.8%)
Mechanic (elastic compression stockings)	1 (0.98%)	0 (0%)	1 (1.2%)
Length of use				.043
1-3 weeks	2 (2.0%)	1 (6.3%)	1 (1.2%)
4-6 weeks	94 (92%)	14 (88%)	80 (93%)
>6 weeks	1 (0.98%)	1 (6.3%)	0 (0%)
Other	5 (4.9%)	0 (0%)	5 (5.8%)
Morphine use				.40
Rescue medication	58 (57%)	11 (69%)	47 (55%)
Standard/scheduled medication	35 (34%)	4 (25%)	31 (36%)
Patient controlled analgesia	6 (5.9%)	0 (0%)	6 (7.0%)
No	1 (0.98%)	0 (0%)	1 (1.2%)
Other	2 (2.0%)	1 (6.3%)	1 (1.2%)
Continuous passive motion device, use	2 (2.0%)	0 (0%)	2 (2.3%)	1.0
Physical therapy				.15
On the same day	85 (83%)	16 (100%)	69 (80%)
Not the same day if surgery takes place later on the day	16 (16%)	0 (0%)	16 (19%)
Other	1 (0.98%)	0 (0%)	1 (1.2%)
Hb check				.001
Always	66 (65%)	4 (25%)	62 (72%)
Only if indicated	36 (35%)	12 (75%)	24 (28%)
X-ray				.54
Direct postoperative	99 (97%)	15 (94%)	83 (97%)
During follow-up visit	1 (0.98%)	0 (0%)	1 (1.2%)
Only if indicated	1 (0.98%)	0 (0%)	1 (1.2%)
Other	2 (2.0%)	1 (6.3%)	1 (1.2%)

Discrete variables are given as frequency (percentage), and continuous variables as mean ± standard deviation (range); bold values indicate statistically significant difference (P < .05).

^aBoth groups of orthopedic surgeons (performing both inpatient and outpatient joint arthroplasty vs only inpatient joint arthroplasty) estimated their average LOS for the standard inpatient total joint pathways.

^bMultiple answers were allowed. Proportions of respondents that made use of this feature were displayed.

Although outpatient pathways more frequently incorporated opioid-sparing regimes in postoperative pain management, no statistically significant differences were observed between pathways (∼70% vs 55%, P ≥ .40) (Table 6, Table 7).

Postoperative checks

Outpatient and inpatient pathways for both hip and knee arthroplasty procedures significantly differed in regime for postoperative hemoglobin (Hb) check (P < .001 and P = .001, respectively) (Table 6, Table 7).

Discussion

Optimization of clinical pathways and logistics led to the introduction of OJA. This study aimed to find protocollary differences between outpatient and inpatient joint arthroplasty pathways. Only slight differences were found in both the intraoperative and postoperative phases and probably relied mainly on logistical grounds to ensure efficient same-day discharge. In addition, we found no significantly different LOS for “standard” inpatients in both groups (orthopedic surgeons who performed both inpatient and outpatient vs inpatient alone), suggesting that many existing inpatient pathways might not be far removed from OJA, and OJA is most likely underutilized nationally [12].

Preoperative characteristics

Preoperative patient education is not only mandatory to obtain informed consent but also important as it can serve multiple purposes [13]: (1) Adequate information may motivate patients to actively participate, adhere to the protocol, and increase their sense of responsibility for a successful surgery [14]; (2) manage and modulate expectations as this is associated with patient satisfaction [15]; and (3) reduce preoperative anxiety related to the upcoming surgical event [14,16]. The format of education can range from conventional (verbal one-to-one/group sessions, written) to more innovative methods (audiovisual, decision aids, and eHealth). Interestingly, we found that informative brochures (94%) and verbal education (∼75%) were the most common used forms of preoperative education, whereas the use of decision aids (∼40%) and eHealth (∼30%) was relatively scarce. Although OJA might be considered more innovative than standard inpatient pathways, no significant differences in the use of innovative preoperative education types were found between the 2 pathways.

In outpatient pathways, respondents were more likely to use (self) predefined selection criteria. This makes sense as, to date, evidence suggests that OJA should be reserved for selected patients [9], and therefore, selection criteria were formulated by various research groups [17,18]. However, interestingly we found that, except for ASA classifications, specific screening tools [19] were hardly used in both pathways.

Although not statistically significant, prehabilitation was more frequently applied in outpatient than in inpatient hip and knee pathways. Evidence suggests that preoperative health status is a strong predictor of good postoperative outcome following joint replacement, and as function often tends to deteriorate in the preoperative period, surgeons have the opportunity to optimize patient’s health status in preparation for surgery. Although prior systematic reviews found conflicting effectiveness of prehabilitation, recent studies found that preoperative interventions (ie, exercise/educational) significantly improved pain perception, function/quadriceps strength, and reduced LOS [20]. We believe that the concept of prehabilitation can be broadened beyond functional improvement alone and should also focus on other components (eg, management of comorbidities, nutritional status, and addressing mental health problems such as depression/anxiety and negative coping strategies) [21].

Intraoperative characteristics

Spinal anesthesia was most frequently used in both outpatient and inpatient pathways. Traditionally, spinal anesthesia has been advocated over general anesthesia. However, high-quality studies comparing these anesthesia types in patients undergoing fast-track or outpatient TJA are lacking. Recent randomized controlled trials by Harsten et al. even failed to demonstrate differences in functional recovery, LOS, urinary complications, and early mobilization, questioning whether the assumed benefits of spinal anesthesia in reducing morbidity is relevant in this specific patient population and a setting with already optimized perioperative care [22]. This suggests that both general anesthetics (ie, modern targeted total intravenous anesthetic and infusions targeted to anesthetic depth) and neuraxial techniques may be used as part of multimodal anesthetic regimens and should be based on institution’s preferences, resources, and skills available [13,23]. In spinal anesthesia, it is important to use a fast-onset and short-duration local anesthetic that advocates early mobilization. Our questionnaire however did not distinguish between the use of local anesthetics in either joint arthroplasty pathways.

LIA was mainly used by respondents performing TKA. A systematic review by Andersen and Kehlet found that LIA was effective in pain reduction, opioid consumption, and LOS, only in TKA [13,24]. Although not supported by the concept of LIA, findings in a study by Keulen et al. suggested that LIA in TKA might compromise early mobilization and same-day discharge by transient common peroneal nerve palsy [6]. Extra cautiousness when injecting LIA in the posterolateral joint capsule was suggested. No undisputed evidence regarding the mixture and dosage of drugs is available, which explains the variability in practices.

Interestingly, PNBs were rarely used by respondents in both pathways for hip and knee arthroplasty. This may be explained by the fact that nerve block was not recommended by the Dutch guidelines and not considered an essential ERAS component by the ERAS Society [13,25]. Although the Dutch guideline has been revised quite recently, it only compared LIA with femoral nerve blocks for TKA. In outpatient pathways, surgeons might have been afraid that blocks interfered with early ambulation, physical therapy, and same-day discharge [26]. With growing evidence demonstrating the effectiveness of “pure sensory” PNBs such as adductor canal block, we might see a change in (Dutch) practice within the next decade [27,28].

The PL approach in THA was most commonly used by our respondents. This is consistent with previous findings [29]. However, the last decade, the DAA gained popularity among orthopedic surgeons, as it is considered muscle-sparing and thought to be associated with less pain, a faster recovery, shorter LOS, and reduced risk of dislocations. This might explain why respondents who perform OJA are more likely to use the DAA. Selection bias may also play a role given the technically challenging nature of the DAA, favoring patients that are also suitable for OJA. However, systematic reviews and meta-analyses are conflicting and, to date, fail to prove the superiority of DAA over other surgical approaches [30,31].

The MMP approach has been the standard in most cases of TKA. Interestingly, none of the respondents preferred minimally invasive surgery approaches (eg, mini-MMP/subvastus/midvastus). Evidence suggests that minimally invasive surgery in TKA might result in a faster functional recovery (and a better early range of motion) but was also associated with wound-healing problems and infections [32].

Although not statistically significant, tourniquet-assisted TKA was less common in outpatient pathways than in inpatient pathways. Evidence suggests that tourniquet-assisted TKA does not reduce total blood loss and may impair early postoperative recovery by pain, swelling, and compromised quadriceps function. A recent study also showed that the use of a tourniquet was associated with higher postoperative pain and an increased risk of serious adverse events (thromboembolic events, infection, and reoperation) [33].

Strong evidence exists on the effectiveness and safety of TXA use in TJA to minimize intraoperative blood loss and transfusion rates [34]. However, significant differences in the regimes (ie, route of administration) were found between outpatient and inpatient pathways. Studies found that all routes of administration (oral, intra-articular, and intravenous) were significantly effective compared with routine hemostasis, and no differences between administration routes were found [35]. A combined administration regimen of TXA methods, however, is supposed to be superior to single administration [36]. The latter was used more often in outpatient hip and knee arthroplasty pathways.

Results showed that outpatient pathways more often included 2 doses of antibiotics, whereas inpatients pathways more often included 3 doses (24-hour span). A recent study found that even single-dose antibiotic prophylaxis is as effective in reducing the risk of periprosthetic joint infections, suggesting that the reduction of antibiotic doses can be safely done in order to optimize pathway logistics [37].

Postoperative characteristics

No significant differences were found between both pathways in terms of opioid use and antithrombotic prophylaxis preference and duration. Interestingly, we found relatively high proportions of standard/scheduled opioids or patient-controlled analgesia in both pathways (relatively more in inpatient than outpatient), and this seems very contradictory to the well-founded opioid-sparing multimodal anesthetic regimen for fast-track and OJA. We advise multidisciplinary teams to address this feature in order to optimize their pathway [13].

The Dutch national guidelines advised thromboprophylaxis with a low-molecular-weight heparin for a period of 4 to 6 weeks after THA and TKA [38,39]. However, guidelines have been developed based on outdated studies with treatment algorithms different from the fast-track methodology. More recent studies suggested a more patient-tailored approach as the incidence of thromboembolic events were substantially lower in a fast-track setting [40]. A recent study even found that “in-hospital only” prophylaxis for patients staying less than 5 days in the hospital was safe [41]. In terms of antithrombotic agent, more recent literature even advocates ASA monotherapy in certain groups of patients [[42], [43], [44]].

Consensus exists patients should be mobilized as early as possible postoperatively to achieve early discharge [13,45]. We found that in ∼20% of inpatient pathways, mobilization was delayed to the next day if surgery took place later that day. We hypothesize that logistics (eg, the reduced availability of physical therapists and nurses during after-work hours) and proper titration of anesthesia are main reasons for this delay and is rarely a problem in outpatient pathways because patients will be mainly operated first or second of the day [6]. Also, for a well-established inpatient joint arthroplasty pathway, we strongly recommend the multidisciplinary team to optimize logistics so that all patients are mobilized by the physical therapist on the day of the operation (also after work hours), regardless of the planned operating time [13,45].

Significant differences in postoperative regime for Hb-check were found between outpatient pathways (check only on indication) and inpatient pathways (standard check). Because of the standard use of TXA in present perioperative care, the blood loss and transfusion rates are dramatically reduced. Therefore, patients—that are often subject to strict selection criteria in OJA—that underwent uncomplicated TJA probably have a negligible risk for transfusion need [46].

Strengths and limitations

To the best of our knowledge, this is the first study that assessed differences between outpatient and inpatient arthroplasty pathways in detail. The results represent the Dutch practice and therefore limits external validity to countries with different healthcare systems or less experience in TJA pathway innovation. While interpreting the results, one should also take into account that guidelines and consensus statements for perioperative care in THA and TKA may be different internationally. The survey was designed by a multidisciplinary team including high-volume hip and knee orthopedic surgeons, researchers, physician assistants, and an epidemiologist. A good sample size was achieved, including surgeons who were a member of either the Dutch Hip Society or Dutch Knee Society (focus groups of the Dutch Orthopeadic Association). Although the overall response rate seems low, the literature shows that the response rate was above average as physicians are often a group with a low survey response rate to web-based surveys [47,48]. Moreover, if we focus at the response rate of our follow-up survey, it is very high at 95%. Yet our results may possess a risk for selection bias in favor of orthopedic surgeons who have more affinity with clinical pathway optimization for TJA.

Conclusions

In conclusion, significant differences between current outpatient and standard inpatient pathways for TJA in The Netherlands were found only in the intraoperative (administration regimes of TXA and antibiotic prophylaxis) and postoperative phase (postoperative mobilization regime in THA, duration of thromboembolic prophylaxis in TKA, and the need for postoperative Hb checks). These differences probably relied mainly on logistical grounds to ensure efficient same-day discharge in outpatient pathways. Our findings also suggested that OJA pathway may be underutilized in the Netherlands and informs orthopedic surgeons and their multidisciplinary team that formulating selection criteria for eligible patients and the logistical optimization of an already successful inpatient pathway may be the key factor for a successful OJA implementation.

Conflicts of interest

B.B. and R.P.M.H. is in the speakers’ bureau of/gave paid presentations for and is a paid consultant for Zimmer Biomet. All other authors have nothing to disclose.

For full disclosure statements refer to https://doi.org/10.1016/j.artd.2024.101343.

CRediT authorship contribution statement

Mark H.F. Keulen: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Validation, Writing – original draft, Writing – review & editing. Yoeri F.L. Bemelmans: Conceptualization, Data curation, Methodology, Writing – review & editing, Writing – original draft. B. Boonen: Conceptualization, Data curation, Investigation, Methodology, Project administration, Supervision, Validation, Writing – original draft, Writing – review & editing, Formal analysis. Roel P.M. Hendrickx: Conceptualization, Data curation, Methodology, Writing – original draft, Writing – review & editing. Ide C. Heyligers: Conceptualization, Data curation, Methodology, Writing – original draft, Writing – review & editing. Martijn G.M. Schotanus: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Writing – original draft, Writing – review & editing.

Supplementary Material. Appendix A: Questionnaire

Section 1: Demographics

• 1.
  What is your age?

  • years
• 2.
  What is your sex?

  • a.
    Male
  • b.
    Female
• 3.
  What type of healthcare facility do you work in?

  • a.
    Academic (university) hospital
  • b.
    Non-academic teaching hospital
  • c.
    Non-academic non-teaching hospital
  • d.
    Private clinic
  • e.
    Other, namely ….
• 4.
  Which of the following arthroplasty types do you perform in your daily practice?

  • a.
    Hip arthroplasty
  • b.
    Knee arthroplasty
  • c.
    Both
  • d.
    Neither
• 5.
  What types of hip/knee arthroplasty do you perform?

  • a.
    Primary arthroplasty
  • b.
    Revision arthroplasty
  • c
    Both
• 6.Approximately how many hip/knee arthroplasties do you perform per year? (In numbers) ……
• 7.
  Do you perform outpatient joint arthroplasty (OJA) of the hip/knee?

  • a.
    Yes
  • b.
    No

If your answer to question 7 was “yes”: The upcoming questions all relate to the outpatient pathway for hip/knee arthroplasty as it is currently being used in your healthcare facility.

If your answer to question 7 was “no”: The upcoming questions all relate to the inpatient pathway for hip/knee arthroplasty as it is currently being used in your healthcare facility.

Section 2: Preoperative pathway characteristics

• 8.
  What type of (pre-)operative education do you use? (Multiple answers are allowed)

  • a.
    Verbal (one-to-one or group meetings)
  • b.
    Informative leaflets
  • c.
    Videos
  • d.
    Decision aids
  • e.
    eHealth (eg, patient journey app)
  • f.
    None
  • g.
    I don’t know
  • h.
    Other, namely …
• 9.
  Should patients undergo preoperative physical therapy and/or exercise programs (also known as “prehabilitation”)?

  • a
    Yes
  • b
    No
• 10.
  Do you use risk stratification aids for preoperative screening? (eg, American Society of Anesthesiologists classification, Charlson Comorbidity Index, protocollary defined selection criteria)

  • a
    Yes
  • b
    No
• 11.
  (If your answer to the previous questions was “yes”) Which risk stratification aids do you use? (multiple answers are allowed)

  • a
    Charlson Comorbidity Index (CCI)
  • b
    American Society of Anesthesiologists (ASA) class
  • c
    Outpatient Arthroplasty Risk Assessment (OARA)
  • d
    Total joint replacement (TJR) risk calculator
  • e
    Readmission Risk Assessment Tool (RRAT)
  • f
    Protocollary defined selection criteria
  • g
    I don’t know
  • h
    Other, namely …

Section 3: Intraoperative pathway characteristics

• 12.
  What type of anesthesia do you prefer?

  • a
    General
  • b
    Spinal with sedation
  • c.
    Spinal
  • d.
    No preference
  • e.
    Other
• 13.
  Do you use peripheral nerve blocks in the multimodal pain protocol for hip arthroplasty? (only hip surgeons)

  • a.
    Lumbar plexus block
  • b.
    Femoral nerve block
  • c.
    Fascia iliaca block
  • d.
    Sciatic nerve block
  • e.
    No
  • f.
    Other, namely …
• 14.
  Do you use peripheral nerve blocks in the multimodal pain protocol for knee arthroplasty? (only knee surgeons)

  • a
    Lumbar plexus block
  • b
    Femoral nerve block
  • c
    Obturator nerve block
  • d
    Sciatic nerve block
  • e.
    Saphenous nerve (adductor canal) block
  • f.
    No
  • g.
    Other, namely …
• 15.
  Do you use “local infiltration analgesia” (LIA) with a long-acting anesthetic?

  • a
    Yes
  • b
    No
• 16.
  (If your answer to the previous questions was “yes”) How is the LIA composed? (multiple answers are allowed)

  • a
    Bupivacaine
  • b
    Ropivacaine
  • c.
    Ketorolac (or another NSAID)
  • d.
    Epinephrine (adrenaline)
  • e.
    Morphine
  • f.
    Clonidine
  • g.
    Corticosteroids
  • h.
    I don’t know
  • i.
    Other, namely …
• 17.
  Which surgical approach do you prefer when performing total hip arthroplasty? (only hip surgeons)

  • a.
    Posterolateral
  • b.
    Direct anterior
  • c.
    Straight lateral
  • d.
    Anterolateral
  • e
    Direct superior
  • f
    No preference
  • g
    Other, namely …
• 18.
  Which surgical approach do you prefer when performing total knee arthroplasty? (only knee surgeons)

  • a.
    Medial parapatellar
  • b.
    Lateral parapatellar
  • c.
    Midvastus
  • d.
    Subvastus
  • e.
    Minimally invasive
  • f.
    Quadriceps-snip
  • g.
    No preference
  • h.
    Other, namely …
• 19.
  Do you administer tranexamic acid (TXA) to prevent perioperative blood loss?

  • a
    Yes
  • b
    No
• 20.
  (If your answer to the previous questions was “yes”) When do you administer TXA? (multiple answers are allowed)

  • a.
    Preoperative
  • b.
    Intraoperative
  • c.
    Postoperative
  • d.
    I don’t know
• 21.
  How do you administer TXA? (multiple answers are allowed)

  • a.
    Oral
  • b.
    Intravenous
  • c
    Postoperative
  • d
    I don’t know
• 22.
  How many doses of antibiotics (prophylaxis) do you usually give perioperatively?

  • a.
    1 (only preoperative)
  • b.
    2 (1 preoperative dose and 1 postoperative)
  • c.
    3 (1 preoperative dose and 2 postoperative)
  • d
    I don’t know
  • e
    Other, namely …
• 23.
  Do you use wound drains perioperatively?

  • a
    Yes
  • b
    No
• 24.
  Do you use a urinary catheter perioperatively?

  • a
    Yes
  • b
    No
• 25.
  Do you use a tourniquet during surgery? (only knee surgeons)

  • a
    Yes
  • b
    No
• 26.
  (If your answer to the previous questions was “yes”) How do you use a tourniquet in primary total knee arthroplasty?

  • a.
    During the entire operation
  • b.
    During cementation
  • c.
    Other, namely …

Section 4: Postoperative pathway characteristics

• 27.
  For inpatients, what is the average length of stay after a primary total hip/knee replacement in your healthcare institution? (in days)

  • …… days
• 28.
  Do you use morphines in the multimodal pain protocol?

  • a.
    Only as “rescue medication”
  • b.
    Standard/scheduled morphine
  • c.
    Patient controlled analgesia
  • d.
    No
  • e
    Other, namely …
• 29.
  Do you treat patients postoperatively with thrombosis prophylaxis?

  • a.
    Yes
  • b.
    No
• 30.
  (If your answer to the previous questions was “yes”) Which anticoagulant/treatment do you prefer for thrombosis prophylaxis? (multiple answers are allowed)

  • a.
    Low-molecular weight heparins (dalteparin, enoxaparin, nadroparin)
  • b.
    Novel oral anticoagulants/Direct oral anticoagulants (apixaban, dabigatran, rivaroxaban)
  • c.
    Platelet aggregation inhibitor (acetylsalicylic acid)
  • d.
    Vitamin K antagonists (acenocoumarol, fenprocoumon)
  • e.
    Pentasaccharides (fondaparinux)
  • f.
    Mechanic (elastic compression stockings)
  • g.
    I don’t know
  • h.
    Other, namely …
• 31.
  For how long do you treat patients with thrombosis prophylaxis?

  • a.
    <1 week
  • b.
    1-3 weeks
  • c.
    4-6 weeks
  • d.
    >6 weeks
  • e.
    I don’t know
  • f.
    Other, namely …
• 32.
  Do patients who are undergoing primary total knee arthroplasty receive a continuous passive motion (CPM) device postoperatively? (only knee surgeons)

  • a.
    Yes
  • b.
    No
• 33.
  Are patients always mobilized postoperatively on the day of surgery (by the physiotherapist or nurse)?

  • a.
    Yes
  • b.
    Not if the surgery takes place later on the day
  • c.
    No, patients are always mobilized the next day
  • d.
    I don’t know
  • e.
    Other, namely …
• 34.
  Is the hemoglobin level checked postoperatively?

  • a.
    Yes, always
  • b.
    Only if indicated
  • c.
    No
  • d.
    I don’t know
  • e.
    Other, namely …
• 35.
  Do you take an X-ray postoperatively?

  • a.
    Yes, the same/next day
  • b.
    Yes, at the follow-up in the outpatient clinic
  • c.
    Only if indicated
  • d.
    No
  • e.
    I don’t know
  • f.
    Other, namely …