Surgical drains are commonly used following soft tissue surgical procedures. By removing serum and blood from the wound, drains reduce fluid pockets that can cause discomfort to the patient, delay healing, exacerbate inflammation, or facilitate infection.1,2 Drains evacuate inflammatory mediators, foreign material, necrotic debris, and bacteria. Closed suction (active) drains are postulated to have a lower risk for nosocomial infection, compared with open (passive) drains; prevent skin irritation secondary to maceration by draining fluid; allow for quantification of fluid production; and improve apposition of tissue flaps and skin grafts because of the suction effect on the surgical bed.2
Despite their many benefits, using drains is not without risk. Perhaps the most concerning adverse effect of prophylactic drain placement in a noncontaminated surgical site is the risk of infection. Drain placement automatically converts a clean procedure into a clean-contaminated one.1,2 A drain creates a conduit between the sterile surgical site and the environment and may allow retrograde bacterial migration. Drains cause an inflammatory (foreign body) response, even when constructed of relatively inert materials such as polyethylene, polyvinyl chloride, or silicone rubber. This response decreases resistance to bacterial colonization of the wound.3 Infection risk may be minimized by managing factors such as drain type, size, and duration of placement4; however, these factors must be balanced with drain efficacy and the risks of occlusion and premature removal.
Placement and management of drains have not been well evaluated in veterinary medicine. Anecdotally, clinicians typically make subjective determinations about drainage, type of drain, and duration of drain placement on the basis of personal experience. It has been recommended to remove thoracostomy tubes when fluid production rate is < 2.0 mL/kg/d (0.9 mL/lb/d); however, a retrospective analysis of outcomes found that there was no difference in hospital stay on the basis of rate of fluid production at the time of thoracostomy tube removal.5 The current recommendation for closed suction drains is to remove them as early as possible, on the basis of evidence of decreasing fluid production and a plateau in fluid volume.1 However, no clear standards for rate of fluid production, timing of drain removal, or how to weigh the importance of these variables have been published. There has been no study in the mainstream veterinary literature examining the outcomes following closed suction drainage of subcutaneous surgical wounds in dogs.
The purpose of the study reported here was to evaluate fluid production and factors associated with seroma formation after placement of closed suction drains in clean surgical wounds in dogs. We hypothesized that the use of these drains in clean surgical fields for skin and reconstructive surgery would be associated with a low rate of complications such as subsequent infection or seroma development. Our major goal was to document the rates of fluid production that might be expected following placement of closed suction drains in noninfected sites and identify useful clinical guidelines for drain removal.
Materials and Methods
Case selection—Dogs undergoing clean surgical procedures between 2005 and 2012 were identified, and their medical records were evaluated retrospectively. Dogs were included if a closed suction draina was placed subcutaneously in a clean surgical wound created for the purpose of treating an abnormality other than infection. Dogs were excluded from further analysis if positive results of microbial culture of a wound were reported during the same hospital visit as their surgical procedure. Dogs were excluded from analysis of fluid production data if there was an incomplete or absent record of drain fluid production but were included in remaining data analysis.
Medical records review—Medical records were searched for the following data: patient signalment, weight, type and location of surgery (including the site of drain placement), drain fluid volumes, histologic evaluation and culture results when available, comorbidities, PCV, total protein concentration and albumin concentration within a month of the hospital visit or within a month before surgery, medications administered in the hospital and at discharge, duration of hospitalization, and development of complications associated with the surgical site either during initial hospitalization or noted in the record within the first month after surgery. The duration of drain placement, cytologic evaluation results, PCV, and total protein concentration of drain fluid were recorded when available. To allow direct comparison among individuals, fluid production rate was calculated (mL/kg/h) for different time points; this allowed comparison of absolute and relative fluid production (total fluid volume and fluid production rate at 12-hour time intervals).
Statistical analysis—A commercially available statistical programb was used. Data were not normally distributed, so differences in continuous variables were evaluated with the Mann-Whitney U test or Wilcoxon rank sum test. Categorical variables were evaluated by means of a Fisher exact test and Pearson correlation analysis. Relationships between continuous variables were evaluated by means of bivariate linear regression. Statistical analyses were performed to evaluate any correlation between the drain volumes (mL/kg) at 12, 24, 48, and 72 hours after drain placement and the total drain volume (mL/kg) for each dog versus factors from the medical record. Factors obtained from the medical record were age, breed, sex, weight, histopathologic diagnosis, surgery performed, duration of hospitalization, and occurrence of complications overall, with separate analysis for dehiscence, infection, or seroma; postoperative antimicrobial use; postoperative NSAID use; PCV; serum protein concentration; serum albumin concentration; and drain location (axillary or inguinal), lateral aspect of the thorax (or abdomen) or dorsum, and proximal portion of a limb. These same factors were evaluated for any correlation with the development of complications overall or the specific complications of seroma formation, dehiscence, or infection. Results are reported as median and range. Values of P < 0.05 were considered significant.
Results
Seventy-seven dogs had closed suction drains placed into clean subcutaneous surgical sites between 2005 and 2012. Median age was 10 years (range, 3 to 14 years), and median weight was 32 kg (70.4 lb; range, 6.4 to 54.0 kg [14.1 to 118.8 lb]). Sex distribution included 37 spayed females, 36 castrated males, 2 sexually intact females, and 2 sexually intact males. Mixed-breed dogs were the largest group (n = 30), followed by Labrador Retrievers (12), Weimaraners (4), and ≤ 3 of 19 other breeds. Preanesthetic analyses performed within a month before surgery included PCV in 77 cases (median, 43.7%; range, 19.4% to 56.8%), total protein concentration in 68 cases (median, 6.5 mg/dL; range, 5.1 to 8.4 mg/dL), and albumin concentration in 62 cases (median, 3.8 mg/dL; range, 2.0 to 4.5 mg/dL). Medications administered in the perioperative period included antimicrobials (n = 18 cases) and NSAIDs (52). Postoperative use of analgesics and antimicrobials was not standardized and varied at the discretion of the clinician. No dog in this study received any immunosuppressive medications such as corticosteroids or chemotherapeutics.
Surgical procedures included mass removal (n = 42), mass removal with skin flap (17), amputation (4), mastectomy (4), thoracic wall resection (3), lymphadenectomy (3), liposuction (2), hemipelvectomy (1), and sialoadenectomy (1). Location of surgery and drain placement was most commonly the lateral aspect of the thorax or abdomen (n = 34); other sites included inguinal (18), proximal portion of a limb (10), head and neck (6), axilla (6), and distal portion of a limb (3). Final diagnoses included soft tissue sarcoma (n = 25), lipoma (18), infiltrative and intramuscular lipoma (13), mast cell tumor (10), adenocarcinoma (3), osteosarcoma (3), and adenoma, melanoma, salivary mucocele, hemangiosarcoma, and squamous cell carcinoma (1 each).
Median duration of hospitalization was 2 days (range, 1 to 10 days), and median duration of drain placement was 1.75 days (range, 0.5 to 5 days). Drain volumes were recorded in 68 of 77 cases; median total volume was 105.3 mL (range, 4.5 to 1,611 mL), and median total volume on the basis of body weight was 3.4 mL/kg (1.5 mL/lb; 0.13 to 53.7 mL/kg [0.06 to 24.4 mL/lb]). Drains were removed as planned by the clinician (n = 56), discharged with the dog for later removal by their primary or referral veterinarian (15), removed by the dog (3), removed because of a malfunction with the drain (2), or replaced with vacuum-assisted closure (1).
Mean fluid production rate (mL/kg/h) at various times after subcutaneous closed suction drain placement in dogs that did (n = 14) and did not (36) develop a postoperative seroma.
Citation: Journal of the American Veterinary Medical Association 245, 2; 10.2460/javma.245.2.211
Fluid production (mL/kg of body weight) from subcutaneous closed suction drains during various intervals after surgery in dogs with seroma formation (n = 14) versus dogs without seroma formation (36).
| Interval | No seroma | Seroma | P value |
|---|---|---|---|
| First 12 h | 2.0 (0–8.6) | 3.6 (0.13–17.6) | 0.076 |
| 0–24 h | 3.4 (0.16–15) | 4.5 (0.13–19) | 0.02 |
| 24–48 h | 1.23 (0–5.9) | 4.24 (0.1–18.1) | 0.056 |
| 48–72 h | 1.3 (0–2.66) | 4.63 (1.4–14.6) | 0.016 |
| Last | 0.13 (0–0.22) | 0.16 (0.02–0.89) | 0.036 |
| Total | 3.56 (0–35) | 6.24 (0.13–39) | 0.038 |
Values reported are median (range). To convert to mL/lb of body weight, divide values by 2.2.
Complications occurred in 33 of 60 dogs in the first month after surgery; for 17 cases, follow-up information was not recorded in the medical record. The most common complication was dehiscence (n = 18), followed by seroma (14), infection (4), hemorrhage (1), and rapid local recurrence of neoplastic disease (1); 4 dogs had > 1 complication.
No correlation was found between total or relative (to body weight) volumes of fluid drainage and treatment with antimicrobials or NSAIDs, serum albumin concentration, total protein concentration, PCV, type of surgery, or final diagnosis. A significant (P < 0.001) correlation was identified between seroma formation after drain removal and fluid production rate (Figure 1). Drain fluid production rate was significantly higher in dogs that formed a seroma versus dogs that did not during the following periods: 0 to 24 hours and 48 to 72 hours after surgery (Table 1). Dogs that developed a seroma also produced a significantly greater volume of fluid at the last time point measured before drain removal and a greater total volume of drain fluid. The number of hours the drain was left in place was not significantly (P = 0.1) different between dogs that did (25 hours; range, 0 to 94 hours) and did not (36 hours; range, 4 to 116 hours) form a seroma. Clinicians often based a decision to remove the drain on the number of days it had been in place, rather than the actual fluid production, meaning that drains were often removed in the face of higher fluid production the longer they had been in place. Seroma formation occurred in 7 of 13 (53.8%) dogs that had drains removed when they were still producing ≥ 0.2 mL/kg/h (0.09 mL/lb/h), whereas seroma formation only occurred in 8 of 57 (14%) dogs for which drains were removed when they were producing < 0.2 mL/kg/h (P = 0.009).
Discussion
Significant correlations were identified in this study between seroma formation and the amount and rate of fluid production. One of the primary rationales for placement of drains in surgical wounds is to prevent the accumulation of fluid pockets that may lead to infection, dehiscence, delayed healing, and discomfort.1 Dogs that had greater fluid production at the later time points after surgery often went on to develop a seroma in the present study. Dogs producing < 0.2 mL/kg/h (4.8 mL/kg/d) at the time of drain removal were significantly less likely to form a seroma, suggesting this may be a useful clinical benchmark for drain removal. This value is higher than the 2 mL/kg/d (0.9 mL/lb/d) rate described in the veterinary literature for removal of thoracostomy tubes6; however, no increase in hospitalization time was detected in a study5 in which thoracostomy tubes were removed with rates as high as 10 mL/kg/d (4.5 mL/lb/d). It was unclear in the present study whether the lower risk of seroma formation in dogs that had lower fluid production was, at least in some dogs, attributable to adequate duration of drain placement or allowing for fluid evacuation from the surgical bed, or whether these wounds naturally produced less fluid and would have had a satisfactory outcome regardless of drain placement.
Little information regarding drains and their effectiveness for seroma prevention exists in the veterinary literature; however, in a previous study7 evaluating surgical removal of intermuscular lipomas from the thigh region in dogs, 4 of 6 dogs without drains formed a seroma, in contrast to 0 of 5 dogs in which Penrose drains were placed. This suggests that the benefit of drains in removing the unwanted fluid outweighed the detrimental effect of the foreign body reaction they induced.
The effectiveness of closed suction drains for prevention of seroma formation is controversial in the human literature. Authors of a review of closed suction drains used for orthopedic surgery concluded that there is insufficient evidence of effectiveness to routinely recommend their use.8 Another study9 found equivalent results with closed suction drainage versus axillary padding after axillary lymphadenectomy or with meticulous subcutaneous suture closure versus a drain after cesarean section in obese women. In 1 study,10 although suture closure and closed suction drainage were found to have equivalent occurrence of complications, both had a superior outcome, compared with control patients who received neither a drain nor meticulous subcutaneous suture closure.
Timing of drain removal, both with regard to duration of placement and quantity of fluid produced, is important in clinical decision making but is controversial. In the present study, the duration that drains were in place was not associated with seroma formation. Although the association was nonsignificant, drains were often removed in the face of higher fluid production after they had been in place for longer periods, either because of concern about infection or perhaps in relation to the cost of patient care in dogs that remained hospitalized for drain management. Given the absence of a correlation between drain duration and infection, it is possible that some clinicians remove drains prematurely when dogs may tolerate their presence longer without adverse effects. However, the higher fluid production in some dogs might represent an inflammatory reaction stimulated by the drain itself; the presence of the drain might have predisposed to seroma formation in that individual. A controlled study of dogs with higher rates of fluid production in which drains are removed from some patients but maintained in others would be required to clarify whether leaving the drain in for a longer period is likely to reduce or increase the risk of seroma formation. Likewise, a controlled study of seroma formation in dogs following similar surgical procedures that do or do not include placement of a closed suction drain would be required to determine whether closed suction drains, in general, reduce the risk of seroma formation.
Timing of drain removal after total joint arthroplasty in humans has been recommended as 24 hours after surgery because fluid production decreases substantially and infection rates increase after this time.4 In contrast, drains are placed for much longer periods after soft tissue procedures; a study11 evaluating seroma formation after breast and axillary surgery in women had a mean duration of drain placement of 9 days, with a maximum of 17 days. In a survey of breast surgeons, the volume of fluid produced, rather than a specific duration of drainage, was identified as the main reason for drain removal, with drains being removed when fluid production reached ≤ 30 mL in a 24-hour period.12 A study13 comparing early removal of mastectomy drains at 48 hours versus drain removal at either ≤ 30 mL in 24 hours or postoperative day 14 was discontinued early because of the markedly higher number of seromas requiring needle aspiration, drain reinsertions, and doctor visits in the group with drain removal at 48 hours. Our observation of the increased risk of seroma formation in dogs producing > 0.2 mL/kg/h at the time of drain removal suggests a similar situation in canine patients.
Placement of closed suction drains in this study was associated with few complications aside from premature removal by the dog (n = 3) and drain malfunction resulting in removal (2). Four of 60 (6.6%) dogs for which there was follow-up information had surgical site infection, a rate slightly higher than the reported ranges in dogs and cats undergoing clean contaminated surgery, independent of whether drains are placed (3.5% to 5.0%).14 Drain placement may have been related to development of infection in these cases, but this cannot be confirmed by the present study. In a recent study by Szabo et al,15 4 of 10 healthy dogs with closed suction abdominal drains developed bacterial contamination of the drain by day 7; however, positive results of culture of the drain tip are of uncertain prognostic value regarding the likelihood of a surgical site infection. In a study16 of 110 human total hip arthroplasty patients with closed suction drains, microbial culture of the drain tip was performed at the time of drain removal; none of the 8 patients with a positive culture result developed any evidence of infection with a mean follow-up time of 2.8 years. Many studies in human medicine have failed to detect a significant difference in infection rates between procedures with and without drains. In a study17 comparing patients with and without closed suction drains after exchange of a temporary tissue expander for a permanent breast implant, there was no difference in the infection rate between the 1,495 procedures with closed suction drains versus the 951 procedures without a drain. In the present study, other factors that may have led to an increased risk of infection included extended anesthetic and surgical times attributable to procedures being performed in a teaching institution. Comparison of infection rates in a population of dogs that did not receive a surgical drain would be warranted to clarify the role of drain placement in the development of surgical site infections.
Limitations of this study included its retrospective nature and therefore variability in treatments and completeness of medical records. Frequency of drain emptying and drain care were not standardized; similarly, data regarding drain fluid characteristics such as cytologic findings and PCV and total protein concentration were variably recorded.
Prospective studies are indicated to compare efficacy of closed suction drains with alternative means of avoiding seroma formation, such as bandages and additional suture closure. In addition, further studies may determine whether maintenance of drains until fluid production rate is < 0.2 mL/kg/h reduces the rate of seroma formation and whether maintenance of drains for longer periods increases the risk of surgical site infection.
References
1. Campbell BG. Bandages and drains. In: Johnston SA, Tobias KM, eds. Veterinary surgery: small animal. St Louis: WB Saunders Co, 2011; 221–230.
2. Dougherty SH, Simmons RL. The biology and practice of surgical drains. Part II. Curr Probl Surg 1992; 29: 633–730.
3. Magee C, Rodeheaver GT & Golden GT, et al. Potentiation of wound infection by surgical drains. Am J Surg 1976; 131: 547–549.
4. Drinkwater CJ, Neil MJ. Optimal timing of wound drain removal following total joint arthroplasty. J Arthroplasty 1995; 10: 185–189.
5. Marques AI, Tattersall J & Shaw DJ, et al. Retrospective analysis of the relationship between time of thoracostomy drain removal and discharge time. J Small Anim Pract 2009; 50: 162–166.
6. Tillson DM. Thoracostomy tubes. Part II. Placement and maintenance. Compend Contin Educ Vet Pract 1997; 19: 1331–1338.
7. Thomson MJ, Withrow SJ & Dernell WS, et al. Intermuscular lipomas of the thigh region in dogs: 11 cases. J Am Anim Hosp Assoc 1999; 35: 165–167.
8. Parker MJ, Livingstone V & Clifton R, et al. Closed suction surgical wound drainage after orthopaedic surgery. Cochrane Database Syst Rev 2008; 4: 1–50.
9. Classe JM, Berchery D & Campion L, et al. Randomized clinical trial comparing axillary padding with closed suction drainage for the axillary wound after lymphadenectomy for breast cancer. Br J Surg 2006; 93: 820–824.
10. Allaire AD, Fisch J, McMahon MJ. Subcutaneous drain vs. suture in obese women undergoing cesarean delivery. A prospective, randomized trial. J Reprod Med 2000; 45: 327–331.
11. Boostrom SY, Throckmorton AD & Boughey JC, et al. Incidence of clinically significant seroma after breast and axillary surgery. J Am Coll Surg 2009; 208: 148–150.
12. Phillips BT, Wang ED & Mirrer J, et al. Current practice among plastic surgeons of antibiotic prophylaxis and closed-suction drains in breast reconstruction: experience, evidence, and implications for postoperative care. Ann Plast Surg 2011; 66: 460–465.
13. Barton A, Blitz M & Callahan D, et al. Early removal of postmastectomy drains is not beneficial: results from a halted randomized controlled trial. Am J Surg 2006; 191: 652–656.
14. Brown DC. Wound infections and antimicrobial use. In: Johnston SA, Tobias KM, eds. Veterinary surgery: small animal. St Louis: WB Saunders Co, 2011; 135–139.
15. Szabo SD, Jermyn K & Neel J, et al. Evaluation of postceliotomy peritoneal drain fluid volume, cytology, and blood-to-peritoneal fluid lactate and glucose differences in normal dogs. Vet Surg 2011; 40: 444–449.
16. Petsatodis G, Parziali M & Christodoulou AG, et al. Prognostic value of suction drain tip culture in determining joint infection in primary and non-infected revision total hip arthroplasty: a prospective comparative study and review of the literature. A rc h Orthop Trauma Surg 2009; 129: 1645–1649.
17. McCarthy CM, Disa JJ & Pusic AL, et al. The effect of closed-suction drains on the incidence of local wound complications following tissue expander/implant reconstruction: a cohort study. Plast Reconstr Surg 2007; 119: 2018–2022.
