High-Pressure Water Injection Injuries of the Hand May Not Be Trivial

• Updated on: 2018-04-08
• Read original article here

High-pressure injection injuries of the extremity are potentially devastating and can result in amputation, infection, and severe disability. These injuries represent less than 0.2% of hand traumas. They most commonly occur from accidents or equipment malfunctions in the industrial setting. The type of material injected is the most important determinant of outcome. Injections with paint, paint thinner, diesel fuel, organic solvent, grease, concrete, plastic, animal vaccines, and water have been reported, with grease and paint historically being the most common. Many of these substances are cytotoxic, resulting in tissue necrosis and a severe inflammatory response. Furthermore, the innocent appearance of the entrance wound often erroneously leads to the misconception of a benign process and hence a delay in treatment.

Industrial applications for high-pressure water jets include cleaning, cutting, polishing, and demolition. Due to its inert nature, high-pressure injection injuries from water generate a less severe inflammatory response and are reported to fare much better than injuries involving more toxic chemicals. Some authors suggest that these injuries can be treated more conservatively with nonsurgical management and antibiotics. Current literature regarding water injection injuries to the hand is limited to scattered case reports, and treatment recommendations are not well established. The purpose of this study was to describe the outcomes of a series of patients with high-pressure water injection injuries.

Once institutional review board approval had been obtained, a retrospective cohort study was conducted of consecutive patients with high-pressure water injection injuries to the hand. The authors' institutional administrative database was queried based on the International Classification of Diseases, Ninth Revision, and Current Procedural Terminology codes for injection injuries between January 1, 2000, and May 12, 2016. Patients injured by high-pressure injection materials other than water were excluded. A total of 19 patients with high-pressure water injection injuries to the hand were identified. Medical records, including admission history and physical examination, operative reports, progress notes, and postoperative visits, were reviewed. Data collected included date of injury, time to surgery, type of treatment, age at the time of injury, sex, postoperative complications, date of surgery, laterality of injury, involved digit, hand dominance, medical comorbidities, previous operations on involved hand, antibiotic use, occupation, and whether the injury was work related. Multiple attempts were made by email and phone to reach all patients to conduct a 4-question long-term outcome survey (Figure 1), which queried permanent loss of sensation, permanent loss of motion, permanent chronic pain, and the need for additional surgery. Patients were classified into 1 of 2 groups based on their initial treatment: (1) those who underwent surgical exploration and debridement in the operating room within 2 days of their injury (early surgery group) or (2) those initially treated nonoperatively owing to surgeon discretion or delayed presentation (nonoperative group). There were no specific criteria for the surgeon's decision to proceed with early surgery vs nonoperative treatment. Antibiotic therapy also varied between patients. The primary outcome variable was the presence or absence of complications requiring surgery after initial treatment. For patients in the early surgery group, this referred to any procedures performed after their index procedure. For patients in the nonoperative group, this referred to any subsequent procedure related to a complication of their water injection injury. Secondary outcome variables included permanent loss of sensation, permanent loss of motion, and permanent chronic pain. All data were collected and stored in the Research Electronic Data Capture database of the authors' institution. Standard descriptive statistics were calculated, including measures of central tendency (ie, means), variance, and frequencies and proportions. Fisher's exact tests were used to compare the proportions of bivariate variables (eg, occupation, antibiotic use) between treatment modalities. Given the small sample, the authors assumed all data were nonparametric (ie, not normally distributed). Thus, a Wilcoxon rank sum test was used to compare averages of continuous variables (eg, age) between groups.

The authors identified 19 patients with high-pressure water injection injuries to the hand or finger, treated by 10 different surgeons, who met their search criteria. Average age at the time of injury was 44 years (range, 23–69 years), and 79% were men. The index finger was the most commonly injured digit (n=9), followed by the middle finger (n=4), the thumb (n=4), the ring finger (n=1), and the volar palm (n=1). Eighty-four percent of injuries involved the nondominant hand. None of the patients had previous surgery on their injured hand. Forty-two percent of the injuries were work related. The exact water pressure gradient was documented for 6 patients, ranging from 2000 to 6000 psi. Nine of 19 patients met criteria for the nonoperative group (Table 1). Average time from injury to initial presentation to a medical facility was 4.2 days (median, 0; range, 0–27 days). Two of the patients in the nonoperative group presented in a delayed manner, 1 at 11 days after injury and 1 at 27 days. Neither received antibiotics immediately after injury. The other patients in the nonoperative group received antibiotics immediately after injury (6 orally, 1 intravenously). Three patients (33%) in the nonoperative group ultimately had complications requiring surgery, including the 2 patients with delayed presentation. The first was a 53-year-old water treatment plant worker who presented 11 days after injury after developing an abscess over the distal aspect of his injured middle finger. He underwent irrigation and debridement (I&D) in the clinic and was started on oral antibiotics. Three days later, he required a more extensive debridement in the operating room for flexor tenosynovitis. Cultures grew methicillin-sensitive Staphylococcus aureus, Citrobacter freundii, and Serratia species. He required flexor tenolysis and manipulation for postoperative stiffness 6 months later. The second patient was a 46-year-old man who presented to the authors' clinic 3 weeks after sustaining a pressure washer injury to his index finger. He had fingertip necrosis treated with debridement and index-to-long cross-finger flap. Ultimately, he underwent revision amputation to his fingertip 1 year later because of progressive necrosis. The third patient was a 58-year-old man who sustained a pressure washer injury to his middle finger. Initial conservative management with oral antibiotics failed. He developed a rapidly progressive infection extending into the palm with flexor tenosynovitis and diminished blood flow and sensation to the middle finger. He underwent I&D 4 days after injury and repeat I&D 2 days later, ultimately requiring amputation of the middle finger at the metacarpophalangeal joint. Summary of Data for Patients in the Nonoperative Group The 10 patients of the early surgery group underwent operative intervention as part of their initial management (Table 2). Average time from injury to operating room was 0.3 days (median, 0; range, 0–1 days). All received antibiotics at initial presentation to the emergency department (9 intravenously, 1 orally). Two patients (20%) treated operatively for their high-pressure water injection injuries had complications requiring additional surgery. The first patient was a 30-year-old man who injured his nondominant middle finger with a metal-cutting machine. He underwent I&D the day of injury and trigger finger release because of scar tissue formation 10 months later. The second patient was a 23-year-old man who injured the palmar aspect of his nondominant hand with a pressure washer. He underwent emergent I&D, carpal tunnel release, and fasciotomies for compartment syndrome of the hand. His ring finger lumbrical muscle was destroyed, and the wound was contaminated with sand. He required 2 additional I&Ds. Intraoperative cultures grew Pseudomonas. Summary of Data for Patients in the Early Surgery Group Nine of 19 patients completed the long-term follow-up survey (Table 3). One of 3 in the nonoperative group had permanent loss of sensation, loss of motion, and pain. The other 2 had no long-term sequelae. Three of 6 in the early surgery group had permanent loss of sensation, 4 had permanent loss of motor function, and 0 had permanent pain. Statistical analysis showed that patients in the early surgery group (10 of 19; 53%) were significantly younger than patients in the nonoperative group (9 of 19; 47%) (median age, 39 vs 49 years, respectively; P=.0410). No significant differences were found between groups regarding sex (P=.9056), type of occupation (P=.9372), whether the injury was work related (P=.1698), affected hand (P=1.000), or time to treatment (P=1.000). Patients in the nonoperative group were significantly more likely to be treated with oral antibiotics than with intravenous antibiotics (P=.0037).

High-pressure injection injuries of the hand are often benign in appearance (Figures 2–3). They typically occur in the non-dominant hand of young working males attempting to clean a nozzle. The index finger is the most common digit involved, followed by the thumb. The palm and the long finger are also commonly injured. The anatomic distribution of injury site in the current patient cohort is consistent with that previously reported. Eighty-four percent of injuries involved the nondominant hand. The index finger (47%) was the most commonly involved digit, followed by the thumb (21%) and the middle finger (21%). High-pressure injection injury to the hand showing the often benign appearance. Close-up of a high-pressure injection injury to the hand showing the often benign appearance. Water jet velocities can reach 900 mph and up to 55,000 psi, resulting in high-energy shock waves that may produce an injury pattern similar to that of a high-velocity ballistic and capable of cutting through concrete, steel, and protective equipment. Most industrial water blasters operate at 8000 to 12,000 psi, whereas most nonindustrial handheld pressure washers operate at approximately 2000 psi. Home pressure washers are commonplace but much less likely to result in serious injury. The exact pressure gradient was documented for 6 of the 19 current patients, ranging from 2000 to 6000 psi. An infection rate of 20% to 35% has been reported for high-pressure water injuries. These injuries may lead to exposure to unique bacterial organisms, depending on the water source. Local water sources such as rivers, lakes, or ponds are sometimes used. Many industrial high-pressure cleaning systems use wastewater and may expose the injured to a variety of organisms, including Aeromonas hydrophila, Corynebacterium aquaticum, Enterobacter cloacae, Pseudomonas aeruginosa, Mycobacterium marinum, and Vibrio species. Skin flora such as Staphylococcus aureus and Streptococcus pyogenes remain the most common organisms. In the current study, 3 (16%) of 19 patients became infected after their water injection injury, with all 3 having negative long-term sequelae as a result. Bacteria cultured from the current patients included methicillin-sensitive Staphylococcus aureus, Citrobacter freundii, Serratia species, and Pseudomonas. Additionally, the patient with Pseudomonas infection had concomitant contamination with sand, which may have influenced his clinical outcome. Amputation is a potentially devastating sequela of high-pressure injection injuries. Hogan and Ruland analyzed 435 cases of high-pressure injection injury to the upper extremity and cited an amputation rate of 40% for paint, 55% for paint thinner, 17% for grease, and 72% for diesel. No amputations were reported in the 11 water injection injuries. In fact, no previous reports of amputation resulting from water injection injury to the hand could be found in the current authors' comprehensive literature search. Two patients from the current cohort, both in the nonoperative group, went on to eventual amputation, including a distal tip revision amputation and a metacarpophalangeal disarticulation. One of these patients presented 27 days after injury, which potentially contributed to his poor outcome. There is little literature to guide decision making for water injection injuries. Hogan and Ruland cited successful operative and nonoperative treatment of high-pressure water injuries to the upper extremity with expected full recovery of motor function and range of motion. Multiple other reports in the hand literature detail successful conservative management. Others advocate conservative treatment unless there are signs of compartment syndrome or the presence of extensive subcutaneous emphysema and systemic symptoms. Operative treatment of water injection injuries to the hand has also been described. Christodoulou et al presented 2 high-pressure injuries to the hand treated with surgical debridement, and Costas-Chavarri et al described surgical debridement and repair of a ruptured index finger flexor digitorum profundus tendon and A3 pulley.

Several key findings can be elucidated from this study. First, this represents the largest published cohort of both operatively and nonoperatively treated high-pressure water injection injuries to the hand. Although involving only 19 patients, this is a sizable case series compared with previous studies and adds to the understanding of this uncommon injury. Additionally, this is the first report of amputation as a complication from water injection injury, which occurred in 2 patients. Compartment syndrome was also a unique complication seen in 1 patient. Approximately 1 in 6 patients developed infection, consistent with previous literature, and this portends a poor prognosis. Contaminants, such as sand, may be present and increase the likelihood of infection. The current study's data are insufficient to provide specific antibiotic recommendations or indicate whether intravenous antibiotics are superior to oral antibiotics. Twenty percent of patients in the early surgery group had complications requiring additional surgery, indicating that even aggressive early management does not completely safeguard from future postoperative complications and that patients should be closely observed after their index procedure. Delayed presentation is a poor prognosticator, as both patients with delayed presentation in the current study went on to have complications requiring surgery, further enforcing the recommendation for immediate medical attention and early antibiotics. Finally, this analysis of patients' long-term outcomes indicates that permanent loss of sensation, permanent loss of motor function, and permanent chronic pain are not uncommon sequelae. This study had several weaknesses, including its retrospective nature. Only 9 of 19 patients completed the long-term outcome survey, despite multiple attempts to reach all patients. Self-selection may have also influenced the results of the long-term outcome survey. Patients with chronic sequelae from their water injection injury may have been more likely to respond to the survey than those without. Additionally, complications may have been underre-ported if patients sought medical attention elsewhere, although all patients appeared stable at their last documented clinic visit. The decision for early surgery vs nonoperative management was based on surgeon discretion. The authors cannot infer selection criteria for operative vs nonoperative management from the results of this study. Additionally, there was variability in antibiotic management, which further limited the authors' ability to directly compare the early surgery and nonoperative cohorts. For example, most patients in the nonoperative group were treated initially with oral antibiotics, whereas most patients in the early operative group received intravenous antibiotics. On the basis of the information from this study and analysis of existing literature, all cases of high-pressure water injection injury to the hand deserve immediate medical attention, including early antibiotics, tetanus prophylaxis, and surgical consult. The authors recommend a low threshold for overnight observation if early surgical debridement is not performed. Although water lacks the direct toxicity of injection materials such as paint, the combination of factors such as the pressure of injection, the depth of penetration, the presence of contaminants, and the delay in presentation contributes to the potential for serious complications. The use of I&D in the operating room is at the discretion of the treating surgeon. This decision must be made on a case-by-case basis and must be based on these factors as well as the patient's medical comorbidities. Patients should be warned of the risk of infection and other long-term sequelae. High-pressure water injection injuries, although often innocuous in appearance, should not be assumed to be benign.