Introduction
Fishhook injuries have been reported in many types of chelonian species and have been identified as a common hazard to sea turtles.1–4 Due to their close proximity with recreational and commercial fishing and the lack of permit enforcement, many freshwater turtles are also at risk of incidental bycatch injuries as well as intentional victims of hooking.1,2,5 In radiographic studies of other wild caught freshwater turtles, up to 36% of individuals in a population were found to have incidentally ingested a fishhook.6,7 It has been suggested that ingestion of these fishhooks can result in a mortality rate of up to 11% if not removed.8
Various treatments and removal techniques have been reported in multiple species of sea turtles, but not extensively described in freshwater turtles. These include an intraoral retrograde approach, removal via endoscopy and esophagostomy, and various approaches for celiotomy to retrieve fishhooks from different aspects of the gastrointestinal tract.9–12 The removal technique used may influence post-hooking mortality, and various factors including the location and type of hook need to be considered when selecting the optimal approach and treatment plan.10 While surgical approaches can be successfully performed with minimally invasive or advanced techniques, fishhook injuries can still result in mortality.5,13 The chronicity of the injury prior to surgical removal should also be considered as it has been shown that turtles can spontaneously expel hooks without intervention.5,13,14 However, when a fishing line is attached to an ingested fishhook, surgical removal is recommended to avoid strangulation and linear foreign body injuries.5
Chelonians have been categorized to have one of the highest vertebrate extinction rates, with over half of the 360 living species being threatened with extinction.15,16 Due to their long lifespan and low fecundity, threats from pollution and fisheries interactions can have a negative impact on their survival.4,16,17 As turtles play important roles in ecosystem health, conservation efforts aimed at decreasing adult mortality should be encouraged to reduce population declines.15,17 Since fishing injuries are known to affect otherwise healthy turtles, the successful treatment of these injuries may aid in conservation efforts.9
The objectives of this study are to describe the clinical characteristics, treatment, and outcomes of freshwater turtles native to central North Carolina that were presented to a wildlife clinic with fishhook related injuries. We hypothesized that the final disposition of the turtles would be impacted by the location of the fishhook foreign body and that non-surgical removal methods would have the lowest complication rate.18 As recreational fishing is an ongoing and unavoidable problem associated with injuries to chelonians,9 this paper identified different techniques to treat these injuries in freshwater turtles as well as evaluate their success rate and possible risks for complications.
Methods
The records of freshwater turtles with fishhook injuries who presented to the North Carolina State University College of Veterinary Medicine Turtle Rescue Team (TRT) between July 1997 to July 2022 were retrospectively examined. Scanned files of paper medical charts were individually reviewed if fishing hook injury was listed as the presenting complaint in the yearly intake log. If no yearly log was present, all available documents were reviewed for mention of fishing injuries from that year. Records from the years 2020 to 2022 were reviewed using a keyword search of the words “fish” and “hook” from the online record database (RaptorMed version 3.17.34); these 2 years correlated to the time the program transitioned to electronic records.
The minimum inclusion criteria for records in this study were notation of the patient’s species, date of presentation, and presence of a fishhook foreign body on physical examination or radiographs, or known removal prior to presentation. Data were manually extracted from the records by 2 observers (LMH and SLK), and word processing software (Excel version 16.69; Microsoft Corp) was used for collection and evaluation. Patient information recorded included signalment (identification number, species, sex, age), weight, days charted, and location found. Data collected from records included the location of the fishhook, the removal technique, diagnostics performed, medical treatments, sedation protocols, and complications that occurred. Butorphanol was used as a sedative or analgesic in several cases (n = 18/111), however it was not recorded as an analgesic in this study due to lack of evidence of efficacy. Due to the nature of the program’s record system evolving over time and variability in medical record writing, complete data were not available for every patient. Therefore the results of individual variables were recorded out of the total number available for each category.
The removal techniques were categorized as retrograde, esophagostomy, endoscopy, celiotomy, modified disgorger technique, debridement, or no intervention. These removal methods are further defined with descriptions of the procedure in the results section. The removals were also classified as pre-admit removal, self-removal in hospital, death before removal, or released without the hook. Any complications, defined as any pathologic change occurring prior to and including the removal of the fishhook that led to an increase in morbidity or mortality, were recorded. The final disposition was documented as released, euthanized, or died while in rehabilitation.
Statistical analysis
Continuous data were assessed for normality using the Shapiro Wilk test using commercial software (Prism version 9.0; Graphpad Software Inc). Other analyses were conducted with a different commercial software (R version 4.2.2; R Foundation for Statistical Computing). For the mortality and complications by hook location, examinations were done without consideration of whether a patient had multiple hooks present. The analysis conducted was the extension of the Fisher exact test to tables larger than 2x2 with the table in question being hook location by mortality/complications. Where this method yielded significant results, follow-up tests were conducted comparing individual locations against one another in 2 by 2 tables by using the largest group (oral cavity) as a reference category to compare others against. The Fisher exact test was again used in considerations of complications versus mortality and removal method as well as mortality versus injury as ocular or other. Where the conventional Fisher exact test was applied, ORs and CIs are reported.
The Wilcoxon rank sum test was used to compare time in care between patients that did and did not have their hooks removed in a retrograde manner. A logistic regression model was fit with final outcome as the dependent variable and hook location (as oral cavity, esophagus, or other), removal method (as retrograde, esophagostomy, removed by finder, or other), sex (as male, female, juvenile, or unknown), species, and number of days in care as independent variables. Backward selection via AIC was then applied. A binomial test was applied to determine whether the breakdown of sex into male or female was in line with our expectations of equal representation. A Clopper-Pearson CI is provided. Statistical significance was designated by values of P ≤ .05.
Results
Patient population
Between July 1997 and July 2022, 126 freshwater turtles presented with fishhook injuries. Yellow-bellied sliders (Trachemys scripta scripta) were the most prevalent species seen (n = 69/126 [54.8%]; 95% CI, 45.7 to 63.6), followed by river/Florida cooter species (Pseudemys spp) and common snapping turtles (Chelydra serpentina; Table 1). Sex was described for 105 of these turtles, with the majority of the affected being female (n = 61/105 [58%]; 95% CI, 48.1 to 67.7). However, females were not significantly more likely to appear than males (P = 0.118). Weight was recorded for 114 of the turtles, and the median weight was 1.57 kg (25% to 75% IQR; 0.55 to 1.87 kg; minimum to maximum: 0.005 to 13.2 kg). While not all records from TRT were available for review, the percentage of known fishhook injuries accounted for 1.8% (n = 126/7132; 95% CI, 1.5 to 2.1) of the total patients evaluated during the duration of this study.
Species of turtles presenting to the Turtle Rescue Team with fishhook injuries from 1997 to 2022 out of a total of 126 turtles.
| Species | Number (% total) | 95% CI |
|---|---|---|
| Yellow-bellied slider (Trachemys scripta scripta) | 69 (54.8%) | (45.7–63.6) |
| Cooter (Pseudemys spp.) | 26 (20.6%) | (13.9–28.8) |
| Common snapping turtle (Chelydra serpentina) | 11 (8.7%) | (4.4–15.1) |
| Common musk turtle (Sternotherus odoratus) | 6 (4.8%) | (1.8–10.1) |
| Red-eared slider (Trachemys scripta elegans) | 6 (4.8%) | (1.8–10.1) |
| Eastern painted turtle (Chrysemys picta) | 3 (2.4%) | (0.5–6.8) |
| Eastern box turtle (Terrapene carolina carolina) | 2 (1.6%) | (0.2–5.6) |
| Eastern mud turtle (Kinosternon subrubrum) | 2 (1.6%) | (0.2–5.6) |
| Unidentified | 1 (0.8%) | (0.0–4.3) |
| Total | 126 |
Physical examination and diagnostic findings
Physical examinations and treatment plans were recorded in 111 turtles. The recorded anatomic locations in which the hooks were embedded were individually classified as the oral cavity, esophagus, lower gastrointestinal tract, limb, or eye (Table 2). To account for turtles who had multiple fishhooks in different sites, as well as those in which 1 fishhook affected multiple sites (eg, oral and ocular cavities), the locations of the fishhooks were recorded as individual sites, regardless of how many fishhooks were present (n = 140).
The locations in which fishhooks were embedded in the turtles presenting to the Turtle Rescue Team with fishing hook injuries from 1997 to 2022. Six turtles had fishhooks present in 2 locations and 1 turtle had fishhooks in 3 locations. There were 20 fishhooks that were embedded in more than 1 location. The total number of locations in which fishhooks were present is 140.
| Location | Oral cavity | Esophagus | Lower GI tract | Limb | Eye | Total locations |
|---|---|---|---|---|---|---|
| Total number of cases affected | 77 (55%; 95% CI, 46.4–63.4) | 26 (18.5%; 95% CI, 12.5–26.0) | 8 (5.7%; 95% CI, 2.5–10.9) | 3 (2.1%; 95% CI, 0.4–6.1) | 26 (18.5%; 95% CI, 12.5–26) | 140 |
One turtle had the hook lodged in the oral cavity as well as through the interdigital epidermis of her left front foot, counting for 2 locations (Figure 1). Nineteen hooks (19/140 [13.6%]; 95% CI, 8.4 to 20.4) that were embedded in the oral cavity also affected the eye. There were 11/111 (10%; 95% CI, 5.1 to 17.0) turtles with multiple fishhooks present. Of these, 7 turtles had fishhooks in 2 locations and 1 turtle had fishhooks in 3 locations. The most common location was the oral cavity (77/140 [55%]; 95% CI, 46.4 to 63.4). Fourteen (14/111 [12.6%,]; 95% CI, 7.1 to 20.3) of the turtles had fishing line present along with a hook.
Image of a yellow-bellied slider (Trachemys scripta scripta) presenting with a fishhook lodged in the oral cavity as well as the interdigital epidermis of her right front foot. (Image courtesy of R. Carpenter.)
Citation: Journal of the American Veterinary Medical Association 261, 12; 10.2460/javma.23.03.0146
Diagnostics that were performed in these cases included measuring manually read PCVs and diagnostic imaging. No cultures, CBC or serum chemistry panels were performed. PCVs were measured in 11 of the turtles (mean 15.64%, SD 6.78; minimum to maximum: 5% to 30%). Survey digital radiographs were collected for 25 turtles to verify the location and number of hooks present (CPI CMP 200 DR; Communications & Power Industries LLC, and Canon CXDI50G; Canon, Inc). A noncontrast CT (Sensation 16-slice CT; Siemens Medical Solutions) was performed in 1 case involving other foreign material and a large abscess in the cervical region.
Treatments performed
Medical treatments included analgesic therapy in 96/111 (86.5%) of the cases. Analgesics used included morphine (n = 36/111 [32.4%]), ketoprofen (23 [20.7%]), ketorolac (12 [10.8%]), lidocaine (7 [6.3%]) and meloxicam (5 [4.5%]). Sedation was used to facilitate the removal of the fishhooks in 81/111 (72.9%) of the cases. Sedative drugs with analgesic properties including ketamine (n = 47/111 [42.3%]), dexmedetomidine (33 [29.7%]) and medetomidine (13 [11.7%]) accounted for 25/96 (26%) cases which received analgesia. Topical medications (silver sulfadiazine, triple antibiotic ointment, antimicrobial spray) were applied in 40/111 (36%) of the cases and systemic antimicrobial therapy was administered in 60/111 (55.8%) of the cases. Antibiotics administered included ceftazidime (n = 44/111 [39.6%]), danofloxacin (7 [6.3%]), enrofloxacin (6 [5.4%]), clindamycin (6 [5.4%]), oxytetracycline (2 [1.8%]), and amikacin (1 [0.9%]). Multiple antibiotics were administered in 6 of those cases. Due to the presence of an incision or wound, 24/111 (21.6%; 95% CI, 14.4 to 30.4) of the turtles were dry docked for longer than a night and fluid therapy (10 to 30 mL/kg, SC, q 24 h) was initiated in 35 (31.5%; 95% CI, 23.0 to 41.0) cases including those dry docked.
Removal techniques
While there were 140 total anatomic sites affected, there were only 120 different removal methods evaluated as the remaining fishhooks affected more than 1 location at once and were therefore removed with a single removal technique (Table 3).
The number of different hook removal methods per anatomic site of fishhook foreign bodies in 111 turtles diagnosed with fishhook injuries from 1997-2022 at a single wildlife center.
| Retrograde | Esophagostomy | Endoscopy | Celiotomy | Benign neglect | Removed by finder | Self-removed | Disgorger successful | Debriding out of soft tissue | Died with hook present | Released with hook | Total (n=120) | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Oral | 63 | 2 | 0 | 0 | 0 | 6 | 4 | 0 | 1 | 1 | 0 | 77 |
| Esophagus | 8 | 12 | 2 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 2 | 26 |
| Lower GI | 0 | 0 | 0 | 1 | 3 | 0 | 0 | 0 | 0 | 1 | 3 | 8 |
| Limb | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 0 | 0 | 2 |
| Eye only | 5 | 0 | 0 | 0 | 0 | 2 | 0 | 0 | 0 | 0 | 0 | 7 |
Retrograde removal
The majority of the fishhook injuries (76/120 [63.3%]; 95% CI, 54.1 to 71.9) were removed in a retrograde fashion, after advancing the hook antegrade to cut the barb off with wire cutters before reversing through the initial puncture wound as needed. This involved grasping the hook with hemostats or another instrument to manipulate it with or without sedation depending on the turtle’s compliance. Eight turtles (8/111 [7.2%]; 95% CI, 4.2 to 13.7) had their hooks removed by their finders, park rangers, recreational fishers, or other veterinarians prior to presentation to TRT with this method. There was no significant relationship between mortality (P = 0.157) or median time and care (P = 0.277) with retrograde removal compared to other removal methods.
Esophagostomy
Overall, 26 (26/111 [23.4%]; 95% CI, 15.9 to 32.4) turtles had hooks embedded in their esophageal mucosa. Esophagostomy was performed to remove the hook in 14/26 (53.8%) of these turtles. This involved either making an incision with a scalpel blade over the palpated hook, pushing the end of the hook through the cervical skin, or performing a cut down with a hypodermic needle adjacent to the hook. A red-rubber catheter was left in place as a drain in 1 snapping turtle (Chelydra serpentina) to allow for flushing of the wound from afar. The skin incisions were closed in an everting pattern. An abscess was present in 5 of these cases at the time of surgery. One turtle had only the barb of the fishhook located during esophagostomy. This was followed by endoscopy to attempt to locate the remainder but was unsuccessful. This patient died post-operatively and post-mortem examination revealed necrotic material in the trachea suggestive of asphyxiation as the cause of death.
Endoscopy
Endoscopy was successfully used in 1 (0.9%) case to visualize the hooks as hemostats were used to grasp them and gently remove them from the esophagus. In another case, endoscopic assisted removal was attempted but the hook was completely embedded in the esophageal mucosa and required surgical removal (ie, esophagostomy). Endoscopy was used as a diagnostic tool to evaluate the integrity of the mucosa of the esophagus after removal in 2 other cases.
Celiotomy
In a snapping turtle (Chelydra serpentina) with a fishhook located in the gastric cavity, a celiotomy was successfully performed and the turtle was released. This was done with a subaxillary approach where the pectoralis and coracobrachialis muscles were bluntly dissected to enter the coelomic cavity. A plastic pipe was placed down the esophagus to help visualize the hooks within the stomach. A small incision was made in the gastric wall at the level of the hook and the barb of the hook was visualized and cut from the shaft. The remainder of the hook and attached fishing line was extracted orally within the plastic pipe. The stomach was closed in an interrupted pattern and the muscle and dermis with a horizontal mattress pattern.
Modified disgorger technique
When fishing line was present or the fishhook was able to be visualized in the oral or esophageal cavity, removal was attempted utilizing an improvised disgorger as described previously in loggerhead sea turtles.11 This involved using 10 Fr and 13 Fr red rubber catheters to thread over the hook with the goal of covering the barb embedded in the soft tissue and slide it out. This was successful in 1 out of 4 cases in which a hard plastic pipe was used instead. While holding tension on the fishing line present, the plastic tube was threaded over the eye of the hook and a notch in the tube was secured over the segment embedded in the tissue. The barb was then manipulated, disengaged, and the hook was removed. In this case, a piece of tissue was still present on the hook which was histologically confirmed to be a piece of chicken gizzard likely used as fishing bait. Endoscopy was used to evaluate the integrity of the esophageal mucosa following this hook retrieval which identified a small tear where the hook was embedded and multiple spots of hyperemia.
Debridement
Three turtles had their fishhooks surgically removed from the soft tissue. One eastern box turtle (Terrapene carolina carolina) had a hook in its right medial tarsus which was easily removed along with necrotic skin and dirt. A snapping turtle (Chelydra serpentina) required removal of tissue from the medial commissure of its oral cavity to extract the hook, and then had metal brackets placed on the maxilla to stabilize it along with wire and suture. A red eared slider (Trachemys scripta elegans) that presented with a fishhook in her oral cavity also had a mass on her distal right forelimb. The lesion was debrided by making an incision over the dorsal and palmar metacarpal joint where a fishhook was found and removed. Postoperative radiographs showed significant osteolysis of carpal bones and ulna, so regional limb perfusion was performed in 4 sessions over 26 days. After a lack of improvement, amputation of the limb proximal to the antebrachium was elected.
No intervention
Four (4/11 [3.6%]; 95% CI, 1 to 9) of the turtles that had ingested the fishhooks passed them with stool prior to release after either no treatment (n = 3/4 [75%]) or administration of oral mineral oil (1 [25%]). Five (4.5%) other turtles were released with the hooks still present within their gastrointestinal tract. These cases were either monitored for movement of the fishhook(s) radiographically with no progression, were eating and defecating normally, or were known to not have recently ingested the hooks according to members of the public (Figure 2). Time in rehabilitation for these cases ranged from 9 to 52 days (median 28.5 days). Three (75%) of these turtles also had hooks lodged in their oral cavity that were removed prior to release.
Dorsoventral full body radiograph of a common snapping turtle (Chelydra serpentina) with metallic foreign objects consistent with 2 fishhooks (A). Recheck radiograph taken 71 days later showed the remaining hook in the cranial coelomic cavity with no significant migration over the recheck interval. The oral hook has been removed prior to this radiograph (B).
Citation: Journal of the American Veterinary Medical Association 261, 12; 10.2460/javma.23.03.0146
Self-removal
Four (4/111 [3.6%]; 95% CI, 1 to 9) of the turtles self-removed the hook from their oral cavity during transport to the clinic (n = 2), after initial sedation for removal (1), or overnight awaiting removal the following day (1).
Ocular related removals
In total, 26/111 (23.4%; 95% CI, 15.9 to 32.4) turtles had ocular related fishhook injuries. Of these, 19/26 (73.1%) of the fishhooks entered through the oral cavity and exited through or around 1 of the orbits and 7 of the fishhooks only impacted the ocular cavities. In 2 of these cases, the hooks entered through the medial canthus of 1 eye and exited through the contralateral medial canthus or globe. Recorded sequela to ocular involvement included ulcerative keratitis (n = 1), anterior uveitis (1), anterior lens luxation (1), and collapsing of the globe (1).
The majority of these cases (19/26 [73%]; 95% CI, 52.2 to 88.4) were managed using the same retrograde technique cited previously, by advancing the hook through the exit location, cutting off the barb if needed, and removing the hook with hemostats. This was done often if there was oral involvement and the hook could be removed without traumatizing the globe, or if the globe was already deemed non-visual. Ocular hooks were removed from 4/26 (15.4%) turtles by the finder prior to presentation, including 1 noted to cause traumatic enucleation or perforation of the globe. In 2 (7.7%) cases, the hooks that damaged the eye worked themselves out during transport to the clinic. In 1 case, a river cooter (Pseudemys concinna) presented with severe right orbital soft tissue swelling, probable globe rupture with regional bone lysis, and active ocular hemorrhage while a fishhook surrounded by caseous material was present further down in the esophagus. This hook was removed via esophagostomy as included previously but is mentioned as it was suspected to cause the ocular disease. There was no significant relationship between mortality and ocular versus other anatomic injuries (P = 0.435).
Patient outcome
Of the 111 turtles with complete medical notes, 99 were released (89.2%; 95% CI, 81.9 to 94.3), 9 died while receiving care (8.1%), and 3 were euthanized after several days of treatment (2.7%). The time from presentation to release ranged from 1 (same day release) to 150 days in care. The median time in care was 16 days (IQR, 6 to 36). There were 5 turtles that were released on the same day as intake, and 30 turtles that were released within a week of presentation. Thirty-six (32.4%) of the turtles received husbandry at an off-site foster home while recovering from their injuries during their rehabilitation course. Three of the 12 turtles that died or were euthanized in care had hooks affecting multiple anatomic locations. Out of the 15 total locations where fishhooks were embedded, 7 (46.6%) were in the oral cavity, 3 (20%) in the esophagus, 1 (6.6%) in the lower gastrointestinal tract, 1 (6.6%) embedded in a limb, and 3 (20%) in the eye. There was no significant difference in mortality by hook location (P = 0.238).
In total, 22/111 (19.8%; 95% CI, 12.9 to 28.5) turtles experienced complications including systemic or localized infection, necrosis of the injury or incision site, and secondary ocular injuries such as corneal ulcers or ruptured globes (Figure 3). Of the 12 deceased turtles, 7 (58.3%; 95% CI, 27.7 to 84.8) experienced a complication. Three of these turtles had an abscess or necrosis present at the site of the fishhook penetration at the time of removal. Two other turtles developed subsequent osteomyelitis or ventral edema surrounding the temporomandibular joint despite being started on antimicrobial therapy upon presentation, and euthanasia was elected due to poor prognosis.
The number of complications by anatomic location of the fishhook foreign bodies in 111 turtles from 1997 to 2022, compared to the total number of fishhooks representing a total of 120 locations.
Citation: Journal of the American Veterinary Medical Association 261, 12; 10.2460/javma.23.03.0146
There was a significant difference in complications by hook location (P = 0.016) attributable to the difference between hooks in the mouth and esophagus (P = 0.02) with complications being significantly more likely in the patients with hooks in their esophagi (OR estimate 3.49; 95% CI, 1.07 to 11.38). However, if the 5 patients were removed who had hooks in the esophagus as well as another location, there was no longer a significant result (P = 0.113). Following logistic regression modeling, we find no interaction effect, but with these 2 terms and using the group with hooks located elsewhere as a reference category we find a significant effect of hooks in the esophagus (P = 0.019; OR estimate 5.61; 95% CI, 1.32 to 23.86) but not the mouth (P = 0.645). There was also a significant relationship between complications and mortality (P < 0.001) where patients who survived were less likely to have experienced complications (OR estimate 0.07; 95% CI, 0.01 to 0.37).
Discussion
In this study we found that the overall release rate of freshwater turtles with fishhook injuries to be high, especially compared to other injuries or illnesses.19 The majority of the turtles presenting were yellow-bellied sliders (Trachemys scripta scripta) likely due to their prevalence as aquatic turtles in this area. There were more females impacted by these injuries than males, however, the ratio of females presented did not differ significantly from an expected 50:50 sex ratio. Fishhook foreign bodies are an uncommon presenting injury, as only 1.8% of the patients presenting to TRT during this study had a fishhook injury. However, the total prevalence is unknown as the sample population does not include turtles that may have died prior to presentation. Due to the nature of the medical record system, the total number of turtles seen by TRT per year could not be accurately determined so it is unknown if these injuries are becoming more or less prevalent over time. Additionally, due to the nature of the record system, the prevalence is out of the total number of patients which includes amphibians and other reptile species.
In the turtles evaluated in this study, the most common treatment administered was analgesic therapy (86.5%). As with many reptilian species, providing effective analgesia in freshwater turtles can be challenging due to the lack of supportive scientific evidence for medications that consistently demonstrate analgesic properties across taxa. Therefore, dosages are often extrapolated from other species and the pharmacokinetics or pharmacodynamics of a given drug may not be available in each species. Additionally, pain recognition in reptiles may be more challenging than mammalian species due to anatomic and physiologic differences.20 Butorphanol was also prescribed to turtles in this study as an analgesic prior to studies showing its lack of efficacy, therefore it was not included in the list of analgesics used.21 Studies evaluating the efficacy of various NSAIDs in chelonian species have also found conflicting results.22 Additionally, the type of COX receptors may differ by species; ball pythons and box turtles have been shown to have an upregulation in COX-1 expression following inflammation while yellow-belly geckos have shown to increase COX-2.23–25 While pharmacokinetic information exists for several NSAIDs in chelonians, it is unknown whether the NSAIDs used in this study resulted in analgesia.26,27 Dosages and durations of medication treatment greatly fluctuated over the years, likely due to clinician preference, changes in protocols and the evolution of reptile analgesic and antimicrobial research. Due to the variability in this medical treatment, the effect of individual medications on outcome was not evaluated. Although this study does not describe pain scoring and evaluation, it is recommended that analgesic therapy should be administered with fishhook injuries according to published dosages.
The most common removal method and location of the hooks was retrograde removal from the oral cavity. This is likely due to the ease and efficiency of removal, as many of the cases seen in this study were released on the same day of admission after this removal technique. Depending on the turtle’s disposition, full sedation may not be required if proper analgesia is administered. However, as some turtles were rapidly released with this method and others, the complication rate in this study may underrepresent the true occurrence as long-term outcome is not available for this released patient population. Due to the overall success rate of retrograde removal, it may still be recommended in uncomplicated cases as the minimally invasive nature could potentially shorten time in care for wild turtles.
Endoscopic removal has been described as the preferred treatment of esophageal fishhook foreign bodies in dogs to decrease complications such as strictures and ischemia.28 It is also reported to be 66% successful in dogs and cats with these injuries and was shown to shorten retrieval time and time in hospital.29 Endoscopy has previously been described in loggerhead (Caretta caretta) and Kemp’s ridley sea turtles (Lepidochelys kempii) as well as a Malaysian giant turtle (Orlitia borneensis).30,31 In this study, endoscopic removal was utilized successfully to aid in the complete removal of a fishhook in 1 case which saw no complications. While there is likely higher success with incisional esophagostomy removals due to the ease of access in freshwater turtles compared to other species, endoscopy may be utilized more effectively in the future to avoid postoperative complications.
While esophageal fishing hooks did not lead to a significantly greater mortality, they were found to be associated with the most complications. Removing the turtles with multiple sites affected impacted the significance of this finding, so the concurrent nature of the injuries influences the severity of complications. However, the second location always included the oral cavity which did not have a high complication rate and is immediately adjacent to the esophagus. There was also a high prevalence of abscesses associated with the injury that were present prior to esophagostomy surgery in these cases. The high risk of complications with esophageal fishhooks may be due to the lack of sterility of the esophagus compared to other organ systems, the difficulty of keeping the site clean and dry during recovery, or the anatomy of turtles necessitating mobility of the injury site which can delay healing. These complications should be avoided when possible due to the significant positive relationship between complications and mortality.
Another technique used to aid in dislodgement of fishhooks in the esophageal mucosa included the use of a disgorger, as previously described in a juvenile loggerhead sea turtle (Caretta caretta).11 This method was noted to have more success with the use of a harder plastic pipe compared to a red rubber catheter and had no reported complications. This technique was also used in this study to aid in the removal of a hook from the gastric cavity in a common snapping turtle (Chelydra serpentina) in combination with celiotomy. A sub-axillary approach was used in this case, which has also been used for gastric fishhook removals in sea turtles as a useful alternative to a plastron osteotomy.5 However, the left pre-femoral approach for a celiotomy has more recently been described to have better visualization of the stomach in snapping turtles (Chelydra serpentina) and requires less muscle dissection compared to the axillary approach.32
Abscess formation at the site of fishhook embedment was a common finding both before and after hook removal, and infection was prevalent in those who experienced mortality. This complication should be considered when creating treatment plans, especially with chronic injuries or a contaminated wound. Due to the cost of obtaining bacterial cultures of the presenting wounds, empirical antibiotic therapy was initiated in the cases evaluated. However, the empirical use of antibiotics should be discouraged, and selection should be based on a culture and sensitivity or Gram stain identification.33 In some circumstances, however, interpretation of culture results may be challenging to differentiate pathogenic bacteria from normal commensals due to the lack of knowledge regarding the normal bacterial flora in reptile species. If critically needed, only first line antibiotics should be used while awaiting the results of a bacterial culture and these should be chosen based on consideration of the common species-specific pathogens.33
A major limitation of this study includes the lack of follow-up after the turtles are released, making it difficult to determine the long-term survivability of the treatment protocols evaluated. As it is also impossible to say if complications such as infection developed in turtles that were released on the same day as presentation, the number of complications associated with different anatomic locations may be underrepresented in this study. Additionally, it is unknown whether long-term complications or mortality were associated with non-removal of internal hooks. Compared to recorded transit times until expulsion in Kemp’s ridley sea turtles (Lepidochelys kempii), the turtles in this study with internal hooks had a longer median time in care, in which the hook would be expected to pass.13 It is also reported that turtles may not pass an ingested hook for up to 2 years, if at all.14 Further studies are needed to determine the long-term effects of benign neglect on intracoelomic fishhook foreign bodies compared to the success of surgical intervention.
Another limitation of this study involves the retrospective nature of reviewing records that contain incomplete descriptions and data. The total number of turtles that presented with fishhook injuries did not have complete medical records available for review, so only the population data were provided for our entire sample size. Diagnostic images were not available for some cases due to transitioning of records and imaging viewers over time. Furthermore, as only 25/111 (22.5%) of the turtles received radiographs in care, the authors cannot be certain that other patients were not released with internal hooks and the prevalence of multiple hooks in the turtles in this study may be falsely decreased. An additional limitation of the retrospective study was the difference in management strategies of cases. As TRT is a student run organization as part of a teaching hospital, case management strategies may have differed between individuals as well as over time. As a nonprofit organization, financial limitations may also affect the length to which some cases were taken.
Several different removal techniques and treatments of fishhooks were described in this study with positive outcomes. Diagnostics such as PCV/total solids may be recommended due to the anemia seen in this study, however only 9.9% (n = 11/111) of animals had PCVs performed and the prevalence of concurrent anemia in fishhook injuries may not be representative due to this small subset of data. Given the comorbidities and risks of infection seen in this study, CBCs and bacterial cultures may also be indicated in fishhook associated injuries. Direct visualization of the hook allowed efficient and uncomplicated removals in many cases; however, based on the removal technique used there was no significant difference in mortality. Fishhook injuries involving the esophagus may require additional interventions to avoid complications including close monitoring, antimicrobial therapy if indicated based on culture and sensitivity, appropriate husbandry, and wound management. Endoscopic exams of the esophageal lining may be beneficial to discover the extent of injury and monitor for healing. Different removal methods such as endoscopy or a modified disgorger technique can also be considered based upon the fishhook injury location. Overall, freshwater turtles with fishhook injuries appear to have a high release rate even when the injuries are severe.
Acknowledgments
The authors have no conflicts of interest to disclose.
The authors would like to thank the student volunteers of the Turtle Rescue Team for providing the care of the patients in this study. We would also like to thank Kent Passingham for guidance with cases and record collection, as well as the Exotic Animal Medicine Service faculty at North Carolina State University for their mentorship on case management.
References
- 1.↑
Donlan CJ, Wingfield DK, Crowder LB, Wilcox C. Using expert opinion surveys to rank threats to endangered species: a case study with sea turtles. Conserv Biol. 2010;24(6):1586-1595. doi:10.1111/j.1523-1739.2010.01541.x
- 2.↑
Rose SA, Bates EB, McNaughton AN, O’Hara KJ, Barco SG. Characterizing sea turtle bycatch in the recreational hook and line fishery in Southeastern Virginia, USA. Chelonian Conserv Biol. 2022;21(1):63-73. doi:10.2744/CCB-1476.1
- 3.
Valente ALS, Parga ML, Velarde R, et al. Fishhook lesions in loggerhead sea turtles. J Wildl Dis. 2007;43(4):737-741. doi:10.7589/0090-3558-43.4.737
- 4.↑
Poli C, Lopez LC, Mesquita DO, Saska C, Mascarenhas R. Patterns and inferred processes associated with sea turtle strandings in Paraíba State, Northeast Brazil. Braz J Biol. 2014;74(2):283-289. doi:10.1590/1519-6984.13112
- 5.↑
Di Bello A, Valastro C, Freggi D, Lai OR, Crescenzo G, Franchini D. Surgical treatment of injuries caused by fishing gear in the intracoelomic digestive tract of sea turtles. Dis Aquat Organ. 2013;106(2):93-102. doi:10.3354/dao02641
- 6.↑
Enge K, Thomas T, Suarez E. Population status, distribution, and movements of the alligator snapping turtle in the Suwannee River, Florida. Florida Fish and Wildlife Conservation Commission. Accessed May 13, 2023. https://myfwc.com/research/wildlife/amphibians-reptiles/turtles/alligator-snappers/suwannee-river/
- 7.↑
Steen DA, Hopkins BC, Van Dyke JU, Hopkins WA. Prevalence of ingested fish hooks in freshwater turtles from five rivers in the southeastern United States. PLoS One. 2014;9(3):e91368. doi:10.1371/journal.pone.0091368
- 8.↑
Steen DA, Robinson OJ Jr. Estimating freshwater turtle mortality rates and population declines following hook ingestion. Conserv Biol. 2017;31(6):1333-1339. doi:10.1111/cobi.12926
- 9.↑
Sivanarayanan TB, Pathak R, Sangeetha P. Intraoral approach for retrieval of fish hook from flapshell turtle (Lissemys punctata). Intas Polivet. 2014;178-179.
- 10.↑
Beenish Q, Jasmeet K, Deepti S, et al. Retrieval of a fish hook from an Indian flap shell turtle. Int J Curr Microbiol Appl Sci. 2020;9(06):1500-1503. doi:10.20546/ijcmas.2020.906.184
- 11.↑
Jaeger GH, Wosar MA, Harms CA, et al. Use of a supraplastron approach to the coelomic cavity for repair of an esophageal tear in a loggerhead sea turtle. J Am Vet Med Assoc. 2003;223:353-355, 311. doi:10.2460/javma.2003.223.353
- 12.↑
Moraes-Neto M, D’Amato AF, Dos Santos AS, Godfrey MH. Retrieval of an esophageal foreign body (fish hook) using esophagostomy in an olive ridley turtle, Lepidochelys olivacea. J Herpetological Med Surg. 2003;13(3):26-28. doi:10.5818/1529-9651.13.3.26
- 13.↑
Alegre F, Parga M, del Castillo C, et al. Study on the long-term effect of hooks lodged in the mid-esophagus of sea turtles. Abstract in: Proceedings of the Twenty-sixth Annual Symposium on Sea Turtle Biology and Conservation. International Sea Turtle Society; 2006.
- 14.↑
Heaton AJ, Pulis EE, Pitchford JL, Hatchett WL, Carron AM, Solangi M. Prevalence and transience of ingested fishing hooks in Kemp’s ridley sea turtles. Chelonian Conserv Biol. 2016;15(2):257-264. doi:10.2744/CCB-1227.1
- 15.↑
Lovich JE, Ennen JR, Agha M, Gibbons JW. Where have all the turtles gone, and why does it matter? Bioscience. 2018;68(10):771-781. doi:10.1093/biosci/biy095
- 16.↑
Stanford CB, Iverson JB, Rhodin AGJ, et al. Turtles and tortoises are in trouble. Curr Biol. 2020;30(12):R721-R735. doi:10.1016/j.cub.2020.04.088
- 17.↑
Heppell SS. Application of life-history theory and population model analysis to turtle conservation. Copeia. 1998;1998(2):367-375. doi:10.2307/1447430
- 18.↑
Stranahan L, Alpi KM, Passingham RK, Kosmerick TJ, Lewbart GA. Descriptive epidemiology for turtles admitted to the North Carolina State University College of Veterinary Medicine Turtle Rescue Team. J Fish Wildl Manag. 2016;7(2):520-525. doi:10.3996/072015-JFWM-056
- 19.↑
Sack A, Butler E, Cowen P, Lewbart GA. Morbidity and mortality of wild turtles at a North Carolina wildlife clinic: a 10-year retrospective. J Zoo Wildl Med. 2017;48(3):716-724. doi:10.1638/2016-0053.1
- 20.↑
Sladky K, Mans C. Clinical analgesia in reptiles. J Exot Pet Med. 2012;21(2):158-167. doi:10.1053/j.jepm.2012.02.012
- 21.↑
Sladky KK, Miletic V, Paul-Murphy J, Kinney ME, Dallwig RK, Johnson SM. Analgesic efficacy and respiratory effects of butorphanol and morphine in turtles. J Am Vet Med Assoc. 2007;230(9):1356-1362. doi:10.2460/javma.230.9.1356
- 22.↑
Ting AKY, Tay VSY, Chng HT, Xie S. A critical review on the pharmacodynamics and pharmacokinetics of non-steroidal anti-inflammatory drugs and opioid drugs used in reptiles. Vet Anim Sci. 2022;17:100267. doi:10.1016/j.vas.2022.100267
- 23.↑
Sadler RA, Schumacher JP, Rathore K, et al. Evaluation of the role of the cyclooxygenase signaling pathway during inflammation in skin and muscle tissues of ball pythons (Python regius). Am J Vet Res. 2016;77(5):487-494. doi:10.2460/ajvr.77.5.487
- 24.
Buch PR, Desai I, Balakrishnan S. COX-2 activity and expression pattern during regenerative wound healing of tail in lizard Hemidactylus flaviviridis. Prostaglandins Other Lipid Mediat. 2018;135:11-15. doi:10.1016/j.prostaglandins.2018.01.002
- 25.↑
Royal LW, Lascelles BD, Lewbart GA, Correa MT, Jones SL. Evaluation of cyclooxygenase protein expression in traumatized versus normal tissues from eastern box turtles (Terrapene carolina carolina). J Zoo Wildl Med. 2012;43(2):289-295. doi:10.1638/2011-0154.1
- 26.↑
Cerreta AJ, Masterson CA, Lewbart GA, Dise DR, Papich MG. Pharmacokinetics of ketorolac in wild Eastern box turtles (Terrapene carolina carolina) after single intramuscular administration. J Vet Pharmacol Ther. 2019;42(2):154-159. doi:10.1111/jvp.12733
- 27.↑
Uney K, Altan F, Aboubakr M, Cetin G, Dik B. Pharmacokinetics of meloxicam in red-eared slider turtles (Trachemys scripta elegans) after single intravenous and intramuscular injections. Am J Vet Res. 2016;77(5):439-444. doi:10.2460/ajvr.77.5.439
- 28.↑
Yardimci C, Inal KS, ÖNYAY T, et al. Interconnected multiple gastroesophageal fish hook retrievals by endoscopic or surgical interventions in dogs: 13 cases (2010-2017). Kafkas Univ Vet Fak Derg. 2020;26. doi:10.9775/kvfd.2019.22766
- 29.↑
Michels GM, Jones BD, Huss BT, Wagner-Mann C. Endoscopic and surgical retrieval of fishhooks from the stomach and esophagus in dogs and cats: 75 cases (1977-1993). J Am Vet Med Assoc. 1995;207(9):1194-1197.
- 30.↑
Pressler BM, Goodman RA, Harms CA, Hawkins EC, Lewbart GA. Endoscopic evaluation of the esophagus and stomach in three loggerhead sea turtles (Caretta caretta) and a Malaysian giant turtle (Orlitia borneensis). J Zoo Wildl Med. 2003;34(1):88-92. doi:10.1638/1042-7260(2003)34[0088:EEOTEA]2.0.CO;2
- 31.↑
Vorbach BS. Use of total parenteral nutrition in the management of an esophageal tear following endoscopic removal of two fishing hooks from a juvenile Kemp’s ridley sea turtle (Lepidochelys kempii). In: Proceedings of the International Association for Aquatic Animal Medicine. International Association for Aquatic Animal Medicine; 2015.
- 32.↑
Jimenez IA, Hermanson JW, Childs-Sanford SE. Surgical anatomy of celiotomy approaches to the stomach in the common snapping turtle (Chelydra serpentina). J Zoo Wildl Med. 2019;50(1):82-88. doi:10.1638/2018-0114
- 33.↑
Hedley J, Whitehead ML, Munns C, et al. Antibiotic stewardship for reptiles. J Small Anim Pract. 2021;62(10):829-839. doi:10.1111/jsap.13402
