The Safety and Efficacy of Using Drainage in Arthroscopic Rotator Cuff Repair

Abstract

Purpose

This study aimed to evaluate the effects of not using a drain or placing a drain in the glenohumeral (GH) or subacromial (SA) joint spaces on fluid retention and pain in the early postoperative period and late clinical outcomes.

Methods

Patients who underwent arthroscopic rotator cuff repair between 2018 and 2020 were included in the study. Before the operation, demographic data, range of motion (ROM), visual analog scale (VAS) scores, Constant–Murley scores has documented. Deltoid muscle diameter (DMD) were measured. The total amount of irrigation used during the surgery and the operation duration were recorded, and the active amount of fluid coming from the drain in patients with a drain was recorded. The first postoperative DMD measure was made in the operating room and accepted as day 0. DMD measurements repeated postoperative first and second day. VAS assessments were performed on the postoperative first and second days. At the outpatient clinic, these measurements were repeated on the first and second weeks after discharge. Functional evaluations were made with ROM and Constant–Murley scores at the final follow-up examination.

Results

There was no difference in the amount of drainage between the two groups in which a drain was used. When the three groups were compared among themselves regarding preoperative and postoperative VAS scores, Constant–Murley scores, and DMD, no significant difference was found.

Conclusions

We do not recommend the routine use of drains after arthroscopic rotator cuff surgery in terms of cost-effectiveness.

Level of evidence

Level II: Prospective Cohort Study.

Introduction

Fluid retention is common finding after arthroscopic shoulder surgery [17]. Especially in rotator cuff tear surgery, working in an area that does not contain a true joint capsule, such as the subacromial(SA) space, increases this risk [16, 17]. In general, it has been reported that it causes problems that do not require additional treatment, such as weight gain of 1–4 kg and soft tissue edema. Rarely, life-threatening complications, such as airway obstruction, respiratory distress, cervical edema, and compression-related neurological injuries, have been reported [2–4, 15, 22].

Although there are authors who place a glenohumeral(GH) or SA drain to reduce fluid retention after surgery, some prefer not to place a drain at all [17]. Many publications in the literature about the use of drains in orthopedic surgery are about arthroplasty and have focused on the knee joint arthroscopy [1, 19–21]. There are not enough studies in the literature regarding postoperative drains after arthroscopic shoulder surgery [10].

The aim of this study was to evaluate the effects of not using a drain or placing a drain in the GH or SA joint spaces on fluid retention and pain in the early postoperative period and late clinical outcomes. We hypothesize that applying a drain after an arthroscopic rotator cuff repair does not affect the clinical results with fluid retention and pain.

Methods

Institutional review board approval was obtained from Gaziosmanpaşa University Clinical Research Ethics Committee (20-KAEK-161). A pre-study power analysis based on previous data determined a sample size of 63 patients (21 per group) to reach the desired power of > 0.8. Deltoid muscle diameter (DMD) increase was the primary outcome measure [8, 15, 22]. Written informed consent was obtained from all patients.

One hundred and twenty-four consecutive patients who underwent shoulder arthroscopy between 2018 and 2020 were enrolled in this prospective randomized study. The patients were randomized preoperatively and divided into three groups: group I, non-drain (ND); group II, patients with an SA joint drain; and group III, patients with a GH joint drain. The randomization sequence generation was obtained from an envelope containing an equal number of ND, SA, and GH group allocation cards drawn in a blinded fashion. A single surgeon who was not blinded to the randomization performed the operations. Patients who underwent SA decompression and acromioplasty with rotator cuff repair to obtain homogeneous groups were included in the study. For the GH joint, patients with Bankart lesions, superior labrum from anterior to posterior tears (SLAP), biceps tenotomy or tenodesis, and isolated subscapularis tendon rupture repair were not included in the study. Seventy-two patients who met the inclusion criteria and continued to outpatient clinic follow-ups were included in the study.

Before the operation, demographic data, range of motion (ROM), visual analog scale (VAS) scores, Constant–Murley scores has documented. Bilateral shoulder DMD were measured. Measurement was made from the most prominent area of the deltoid muscle by marking the lateral distal acromion anteroposteriorly from the axillary line (Fig. 1). All postoperative measurements were made by an independent observer who was not included in the study. The total amount of irrigation used during the surgery and the operation duration were recorded, and the active amount of fluid coming from the drain in patients with a drain was recorded. The first postoperative DMD measure was made in the operating room and accepted as day 0 (Fig. 2). DMD measurements repeated postoperative first and second day. VAS assessments were performed on the postoperative first and second days. At the outpatient clinic, these measurements were performed on the first and second weeks after discharge. Functional evaluations were made with ROM and Constant–Murley scores at the final follow-up examination.

Fig. 1.

Deltoid muscle diameter measurement. a Determination of the measurement location based on reference points, b Taking the measurement

Fig. 2.

Deltoid muscle diameter measurement at the end of the surgery

The beach chair position was used in all patients under general anesthesia. Systolic blood pressure was kept below 100 mm Hg using a hypotensive anesthesia protocol. Saline solutions of 3000 ml each were used at a pressure of 50–70 mm Hg with the aid of an arthroscopic pump(Arthrex®, Continuous Wave™ Arthroscopy Pump, 1370 Florida, USA). Adrenaline 1 ml was added to each 3000 ml saline solution. A 400 ml standard surgical suction drain was used(Bıçakçılar® Redon Dren, 34,522, İstanbul, Turkey).

During the postoperative follow-ups, cold therapy was applied continuously for the first 24 h with a Cryo/Cuff™ Ic Cooler (DJO, 2900 Lake Vista Drive Lewisville, Texas, USA) for the shoulder area. For postoperative analgesia, 100 mg intravenous tramadol three times a day and 10 mg intravenous paracetamol three times a day were ordered to avoid nonsteroidal anti- inflammatory drugs (NSAID) use. The drain was removed within the first 24 h after surgery. Phase 1 exercises were started on the first postoperative day, and the patients were discharged on the second postoperative day. Only passive ROM exercises were initiated in the first 3 weeks, and active assisted ROM exercises were initiated after the third week. After waiting 6 weeks for the repaired tissue to heal, the patients were directed to the physical therapy and rehabilitation clinic. Patients were routinely called for polyclinic follow-ups on the seventh day, second week, sixth week, third month, and sixth month.

IBM SPSS Statistics Software (SPSS Inc., Chicago, IL, USA) version 23.0 was used to analyze the data. The distribution of data was evaluated with the Kolmogorov–Smirnov test. ANOVA was applied to normally distributed data for more than one independent group, and Kruskal–Wallis analysis was applied to non-normally distributed data. The Mann–Whitney U test was used to compare the amount of fluid coming from the drain. The post-hoc Tamhane test was applied to significant parameters with the Kruskal–Wallis test. Repeated ANOVA and the Wilcoxon test were applied for dependent groups. The chi-square test was used to evaluate categorical variables. A p-value < 0.05 was considered statistically significant for all tests.

Results

The mean age of the 72 patients (48 F/24 M) included in the study was 59 years (range: 36–75), and the mean follow-up period was 9 months (range: 7–13). There was no statistically significant difference between the age and gender distributions of the groups. The mean operation time was 123 min, and when the mean operation times between all groups were compared, there was no significant difference between groups. The average amount of irrigation fluid used was 24 L (range 18–36 L) (3000 ml per unit); there was no statistically significant difference between the groups in terms of their averages (Table 1). There was no difference in the amount of drainage between the two groups in which a drain was used. In these groups, drainage during the first 24 h was 69 ml in Group II (SA) and 74 ml in Group III (GH) (Table 1). The distribution of tear types within the groups could not be homogenized and is shown in Table 2.

Table 1.

Distribution of variables between groups

	Group I (ND) (n = 23)	Group II (SA) (n = 24)	Group III (GH) (n = 25)	p
Age	58.5 ± 8.2	60.9 ± 5.4	57.6 ± 9.3	0.324
Sex (M/F)	11/12	4/20	9/16	0.072
Surgery duration (min.)	128.69 ± 27.06	118.33 ± 28.95	122.20 ± 21.75	0.392
Fluid usage (liters)	8.30 ± 1.45	8.25 ± 1.75	7.96 ± 1.42	0.803
Drainage (ml)	No Drainage	68.75 ± 50.67	74.0 ± 56.12	0.748

Table 2.

Distribution of tear types among groups

Type of Tear	Group I (ND) (n = 23)	Group II (SA) (n = 24)	Group III (GH) (n = 25)		Total (n = 72)
Isolated supraspinatus tear	13	16	23	52
Supraspinatus tear with additional tendon injury
Subscapularis	7	3	2	12
Infraspinatus	3	1	0	4
Subscapularis + In fraspinatus	0	4	0	4

The preoperative Constant–Murley mean score was 49 (range: 36–62), while the postoperative mean score was 86 (range: 62–96). The preoperative VAS average was 6 (range: 4–8), while the postoperative final follow-up VAS average was 1.29 (range: 0–5). Within the groups, VAS scores, Constant–Murley scores, DMD, and ROM were compared preoperatively and postoperatively, and a statistically significant improvement was observed in all variables. (Tables 3, 4).

Table 3.

Preoperative and postoperative difference of mean values within groups

	VAS		Constant Murley		DMD (cm)*		p
	Preop	Postop	Preop	Postop	Preop	Postop
Group I (ND) (n = 23)	5.9 ± 1.2	1.4 ± 1.1	49.3 ± 6.0	87.5 ± 5.5	32.3 ± 5.3	33.1 ± 5.3	< 0.001
Group II (SA) (n = 24)	5.8 ± 1.4	1.2 ± 1.1	47.0 ± 7.9	85.7 ± 8.0	33.6 ± 5.4	34.0 ± 5.6	< 0.001
GroupIII (GH) (n = 25)	6.2 ± 1.1	1.1 ± 1.0	50.5 ± 6.5	84.8 ± 5.6	33.7 ± 5.0	34.4 ± 4.9	< 0.001

Values are expressed as mean ± standart deviation

Preop preoperative, Postop postoperative

*Between 0 and 24 h postoperative

Table 4.

Preoperative and postoperative ROM difference of mean values within groups

	Group I (ND) (n = 23)			Group II (SA) (n = 24)			Group III (GH) (n = 25)
	Preop	Postop	p	Preop	Postop	p	Preop	Postop	p
Abduction	96.3	162.1	0.000	94.1	164.5	0.000	92.4	162.4	0.000
Forward flexion	121.9	158.6	0.000	121.6	161.6	0.000	117.4	166.6	0.000
Internal rotation	L4	L2	0.000	L4	L2	0.000	L4	L2	0.000
External rotation	23.9	40.8	0.000	27.7	40.7	0.000	25.8	40.4	0.000

When the three groups were compared among themselves regarding preoperative and postoperative VAS scores, Constant–Murley scores, and DMD, no significant difference was found. Regarding the decreased rates in the DMD change in the first 24 h, although there was a clinical difference between the groups, there was no statistically significant difference (p = 0.185). In contrast to the DMD change rate, there was no clinical or statistical difference in terms of the rate of decrease in VAS score change in the first 24 h (p = 0.603) (Table 5).

Table 5.

Postoperative evaluation of groups

Variables	Group I (ND) (n = 23)	Group II (SA) (n = 24)	Group III (GH) (n = 25)	p
VAS Score
Preoperative	5.9 ± 1.26	5.83 ± 1.40	6.20 ± 1.15	0.587
Postoperative 24 h	4.0 ± 0.76	4.16 ± 1.09	3.92 ± 0.81	0.719
Postoperative 48 h	3.04 ± 0.82	2.91 ± 0.82	2.92 ± 0.86	0.808
Postoperative 1w	1.78 ± 0.67	2.0 ± 0.72	1.96 ± 0.61	0.492
Postoperative 2w	1,78 ± 0.67	1.70 ± 0.55	1.80 ± 0.50	0.832
Constant
Preoperative	49.30 ± 6.05	47.08 ± 7.99	50,560 ± 6.54	0.186
Postoperative	87.52 ± 5.57	85.75 ± 8.01	84,840 ± 5.69	0.223
DMD (cm)
Preoperative	32.30 ± 5.34	33.67 ± 5.45	33.72 ± 5.03	0.508
Postoperative 0 h	37.04 ± 5.63	37.91. ± 5.32	38.40 ± 5.22	0.591
Postoperative 24 h	35.69 ± 5.91	36.87. ± 5.11	36.61 ± 5.32	0.492
Postoperative 48 h	34.69 ± 5.32	35.95. ± 5.37	36.20 ± 5.26	0.517
Postoperative 1 w	32.56 ± 5.27	33.87. ± 5.41	33.84 ± 5.06	0.530
Postoperative 2 w	32.43 ± 5.40	33.66 ± 5.45	33.80 ± 5.00	0.525
Percentage DMD decrease (%)*	3.79 ± 1.96	2.84 ± 1.41	3.19 ± 1.56	0.185
Percentage VAS score decrease (%)*	25.14 ± 20.1	24.05 ± 22.2	26.01 ± 18.64	0.603

Values are expressed as mean ± standart deviation

* DMD and VAS decrease between 0 and 24 h postoperative

Discussion

This study found no statistically significant difference in the late period in terms of pain, fluid retention, ROM, and clinical outcomes in all patients after shoulder arthroscopy.

Fluid retention is more of a problem during shoulder arthroscopy than knee arthroscopy. Increasing the depth of the tissue passed through makes it difficult to reposition the cannulae. Opening a new portal further traumatizes the soft tissues and increases fluid extravasation. Although there is no defined upper limit for the amount of irrigation fluid used during shoulder arthroscopy, studies on symptomatic patients have been reported to be between 20 and 36 L. Therefore, volumes of less than 20 L can be considered safe [17] .For the relationship between operation time and fluid retention, it has been suggested that the maximum operation time should be limited to between 90 and 120 min. [14]. In a study by Lo et al., the average operation time was 91.2 min, and an average of 30 L of irrigation fluid was used [15]. In the present study, the mean operation time was 123 min, and the average amount of irrigation fluid used was 24 L There was no significant difference between all groups in terms of mean operation times and amount of irrigation fluid used. In addition to rotator cuff injuries, surgical procedures like biceps pathologies and acromioplasty, which prolong the operation time, increase fluid retention [2, 3, 15–17, 22]. In this study, patients who underwent additional procedures to achieve standardization were excluded, except for acromioplasty.

In a study evaluating patients who underwent arthropump and hand pump-assisted surgery, it was reported that the mean DMD change rate was 7.27%, with an average measurement change of 2.65 cm, when the preoperative and postoperative 24-h changes were compared [8]. Capito et al. found a mean change in DMD of 2.8 cm using hyperosmolar irrigation fluid and an arthropump [6]. In the present study, when a comparison was made in terms of DMD change rate between hour 0 and 24 h when the drain was removed, a 2.9% change was found in group II (SA) and a 3.2% change in group III (GH). In the first 24 h, a significant decrease in DMD change rate was detected in both groups in which drains were used. In group I (ND), a 3.8% DMD change rate was detected. Although it was not statistically significant, it was found that the use of a drain in group II (SA) caused a reduction in the DMD change rate in the first 24 h. However, when evaluated in terms of VAS scores, the reduction in pain in the first 24 h was not as significant as the DMD change rate. When the percentage decrease in VAS scores was evaluated within the first 24 h, it was observed that four patients in the GH group and two patients in the ND group had a one-point VAS score increase, while all of the other patients had a decrease in VAS scores. However, when an evaluation was made between the groups, there was no statistically significant difference. When evaluated in terms of VAS scores and DMD change rate on the 14th day, there were no clinical or statistically significant differences.

The advantages of intra-articular drains are that they prevent synovitis due to hematoma formation, thus contributing to early ROM, they reduce pain due to joint distension, and they prevent neurological complications due to compression. [5, 11, 23, 25] In addition, they significantly reduce infection and other wound complications that may develop due to hematoma formation. [12] The wound dressing is less contaminated and the dressing frequency decreases, because there is less fluid and blood leaking from the wound. [5, 25] Although hematomas are generally thought to be associated with early postoperative morbidity, it has been shown that they may also adversely affect functional outcomes in the long term. [11, 21] The disadvantages of using drains are foreign body reactions, mechanical problems (entrapment of the drain by being inserted into the tissue by mistake), fluid and electrolyte losses, and increasing wound infections. [23].

Many studies advocate the role of drains in orthopedic surgery. They were also recommended to minimize infection [27]. Tatari et al. stated that the use of drains is necessary for arthroscopic knee surgery [25]. Karahan et al. obtained significantly better ROM and lower VAS scores on the seventh day in patients who had a drain [13]. The present study observed that the use of drains did not affect VAS scores in the early period or functional scores and ROM in the late period. This is thought to be related to the small amount of active drainage that occurs after shoulder arthroscopy, as in knee arthroscopy or arthroplasty, as seen in this study.

Varley et al. used ultrasound to evaluate patients with drains. They suggested that drains can prevent hematoma formation, but only as long as they stay in place. They reported that after drain removal, the hematoma can reach the size it might have been if no drain had been used [26]. It was reported that 85–91% of drainage related to drain use occurs within the first 24 h [9, 21]. They reported that prolonged drainage did not reduce hematoma formation and actually increased the rate of superficial wound infections; 25% of the drain ends were colonized within 48 h [7, 18, 28]. Stevens [24] raised concerns about the use of drains when he reported an increased infection rate for orthopedic surgeries using drains. Studies on the role of drains in total joint arthroplasties have reported that drains are not necessary for wound problems [1, 19, 20]. In the present study, no wound infections were found in the groups with drains. Moreover, no additional complications, such as hematomas, were encountered in the groups with and without drains.

This study has several limitations. It would be more appropriate to carry out the DMD and VAS evaluations every day until the second week. Since no additional postoperative imaging was performed, the exact time of fluid retention from the tissues could not be determined. We chose the inclusion criteria very narrowly and tried to achieve homogenization by forming groups of patients with the same surgical procedure. However, it was not possible to provide homogenization in terms of rotator cuff tears between groups. Similar studies accompanied by imaging methods with larger samples are needed.

This study will contribute to the literature, because it is randomized, prospective, and single-blinded. Furthermore, all operations were performed by a single surgeon, and another observer performed patient clinical evaluations.

Conclusion

As a conclusion, although not statistically significant, this study showed that the use of drains in patients who underwent arthroscopic rotator cuff repair was adequate in terms of regression of early postoperative fluid retention. It also showed that it does not affect early wound complications and pain. It was found that the use of drains in the late period did not make any difference in terms of ROM, VAS scores, and Constant–Murley scores. We do not recommend the routine use of drains after arthroscopic rotator cuff surgery in terms of cost-effectiveness. It does not affect early and late period pain or clinical and functional results.

Compliance with Ethical Standards

Conflict of interest

The authors whose names are listed immediately below certify that they have NO affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, afliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.

Ethical standard statement

This article does not contain any studies involving animals performed by any of the authors. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was carried out after the approval of the local ethics committee (Tokat Gaziosmanpaşa University Clinical Research Ethics Committee).

Informed consent

Informed consent was obtained from all individual participants involved in the study.