• Open Access

Comparison of 2 Retrieval Devices for Heartworm Removal in 52 Dogs with Heavy Worm Burden


  • W.K. Yoon,

    1. Section of Small Animal Internal Medicine, College of Veterinary Medicine and Institute of Veterinary Science, Kangwon National University, Chuncheon 201-100, Korea
    Search for more papers by this author
  • R. Choi,

    1. Section of Small Animal Internal Medicine, College of Veterinary Medicine and Institute of Veterinary Science, Kangwon National University, Chuncheon 201-100, Korea
    Search for more papers by this author
  • S.G. Lee,

    1. Cardiology Service, Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, NC
    Search for more papers by this author
  • C. Hyun

    Corresponding author
    • Section of Small Animal Internal Medicine, College of Veterinary Medicine and Institute of Veterinary Science, Kangwon National University, Chuncheon 201-100, Korea
    Search for more papers by this author

Corresponding author: Changbaig Hyun, DVM, MVSt, PhD, Section of Small Animal Internal Medicine, College of Veterinary Medicine, Kangwon National University, Chuncheon, South Korea 201-701; e-mail: hyun5188@kangwon.ac.kr



For treating dogs with heavy heartworm infection, mechanical removal using various retrieval devices is useful. However, the efficacy and safety of retrieval devices have rarely been studied.


Catheter-based heartworm removal using 2 retrieval devices (basket and tripod grasping forceps) is efficient and safe for treating dogs with heavy worm burden.


Fifty-two client-owned dogs with heavy (Class III and IV) worm burden.


A retrospective study was performed on 52 dogs, using a catheter-based heartworm removal approach using 2 types of retrieval devices (ie, the basket and the tripod grasping forceps). The efficacy and complications associated with the 2 devices were assessed.


The basket device was used on 22 of the study group dogs, and the tripod grasping forceps was used on 30 of the dogs. The postoperative survival rate was 95.5% for the basket device and 80% for the tripod grasping forceps, but the difference was not statistically significant. The worm number captured per attempt was 3.5 ± 1.7 using the basket device and 1.9 ± 0.85 for the tripod grasping forceps (P < .05). Various complications associated with heartworm removal were noticed with both retrieval devices.

Conclusions and Clinical Importance

This study suggests that catheter-based heartworm removal is not only a relatively safe and efficient therapeutic method in dogs with heavy worm burden, but more efficient using the basket device. Our data do not indicate a clear safety advantage between the 2 devices evaluated, although the survival rate was numerically higher in dogs undergoing a basket intervention.


Surgical removal of adult heartworms can be accomplished via either a thoracotomy or jugular venotomy. This approach has been shown to be useful in heavy heartworm infections,[1, 2] particularly in dogs that presented with caval syndrome. Caval syndrome in dogs is a severe variant or complication of heartworm disease (HWD). The most common treatment for dogs with caval syndrome is mechanical heartworm removal (worm embolectomy).[3, 4] Pulmonary thromboembolism and immune-mediated complications (eg, allergic pneumonitis, glomerulonephritis) are the most commonly reported complications associated with adulticide treatment[5] as a result of a marked host immune response against dead worms. Therefore, the mainstays for treating dogs with heavy worm burden are surgical or mechanical heartworm removal, using retrieval devices (ie, worm embolectomy). Although surgical removal from the right atrium or pulmonary vasculature has been attempted previously,[1, 6, 7] it is not the preferred method of treatment because of the invasiveness of techniques and high mortality.[2, 3]

Mechanical removal using retrieval devices (eg, tripod forceps, basket device or a loop snare device) has a successful history in veterinary medicine.[2, 4, 8-12] The main advantages of these methods are decreased invasiveness of the procedure, less damage to the vascular endothelium, and the decreased anesthesia time.[2] Specifically, a catheter-guided technique for worm removal using retrieval baskets and tripod forceps has been successfully performed in dogs with heavy worm burden previously.[10, 12] This technique has been known to improve access to the pulmonary arteries, and to minimize the vascular and intracardiac damage often associated with blind grasping using a retrieval forceps.[10, 12]

Although several retrieval or grasping devices have been used for heartworm removal, the efficacy and safety of these devices have not been assessed. In this study, we evaluated the efficacy and safety of the 2 retrieval devices (basket retrieval device and tripod grasping forceps) most often used by veterinary practitioners for heartworm removal.

Materials and Methods

Study Population and Design

Fifty-two dogs (33 male, 19 female) with heartworm infection (positive by heartworm antigen test1) were referred to the Veterinary Teaching Hospital, Kangwon National University, for interventional heartworm removal between 2006 and 2012. Written consent was obtained from the owners of each dog used in this study. The severity of heartworm infection was graded using the guidelines[3] published by Atkins et al, based on clinical signs and diagnostic imaging studies (assessed by one of the investigators, CH). There were 11 dogs in class III (heavy worm burden) and 41 dogs in class IV (caval syndrome). Interventional extraction was used to treat the dogs in class III with severe clinical signs associated with heartworm disease to avoid postadulticidal complications. The mean body weight and age of the animals in the study population was 12.0 ± 8.9 kg (range: 2.5–45 kg) and 7.7 ± 4.9 years (range: 2–15 years). The animals were divided into 2 groups based on the retrieval device used for heartworm removal. The mean number of worms removed per dog, attempts or time to complete the heartworm removal, and the number of dogs making full recovery (survival rate) were used as variables to compare and evaluate the efficacy of each retrieval device. The mean numbers of macerated worms, type and rate of intraoperative and postoperative adverse events, and total number of intraoperative deaths or fatalities after heartworm removal were additional variables used to compare the retrieval devices.

Methods and Devices Used in the Interventional Heartworm Removal


To minimize the adverse effects of the procedure, the dogs were given prednisolone2 (0.5 mg/kg, PO, q12h), doxycycline3 (5 mg/kg, PO, q12h), and clopidogrel4 (2~4 mg/kg, PO, q24h) 1 week before the procedure, unless the procedure was an emergency. Enoxaparin5 (100 U/kg) was administered SC before the procedure and tramadol6 (2 mg/kg, SC) preoperatively for pain control. The heartworm removal techniques and devices used in this study all have been reported previously.[10, 12]

Heartworm Removal Using a Tripod Grasping Forceps (Tripod)

The dogs were premedicated with atropine7 (0.05 mg/kg, SC) and diazepam8 (0.5 mg/kg, IV). This was followed by induction of anesthesia with propofol9 (4 mg/kg, IV). After tracheal intubation, anesthesia was maintained by isoflurane with a 2–5% concentration depending on the dog's size. Mechanical ventilation was applied at a rate of 30 times per minute using a volume-cycled respirator.10 In dogs weighing <7 kg, after achieving surgical anesthesia, venipuncture was performed at the right jugular vein with an 18G needle. Under fluoroscopy, a guidewire11 was inserted into the needle and then pushed to the cranial vena cava and right atrium; at this point the tip of the guidewire was orientated nose down, and then was advanced into the right ventricle and pulmonary artery. An introducer12 then was inserted into the right external jugular vein with the guidance of the preplaced guidewire. When the guidewire was not able to be directed toward the pulmonary artery, the introducer was inserted into the bottom of the right ventricle via the preplaced guidewire at the right atrium, and then the guidewire was inserted into the introducer to locate the pulmonary artery, and the guidewire was then removed from the jugular vein. In dogs weighing >7 kg, a small incision was made on the exposed jugular vein, and then the introducer was directly inserted into the right ventricle and pulmonary artery. A flexible 3-wire nail-tipped forceps (Tripod; Fig 1A)13 was inserted into the introducer and this was used to grasp the heartworms under fluoroscopic guidance. After several unsuccessful attempts at worm retrieval, the introducer was pulled back to the right ventricle and further worm removal was accomplished. Lastly, the introducer was pulled back into the right atrium and then the worms in the right atrium were removed. The jugular vein was ligated, after the procedure.

Figure 1.

The heartworm retrieval devices used in this study. (A) The flexible 3-wire nail-tipped forceps (Tripod), (B) the basket retrieval device (Basket).

Heartworm Removal Using a Basket Retrieval Device (Basket)

Surgical anesthesia was achieved as described above, and venotomy was performed on the left jugular vein using a surgical blade. The introduction of guidewires and introducers was done as described above. The basket14 (Fig 1B) was inserted into the introducer and later opened to capture and collect heartworms under fluoroscopic guidance at the target area, with a pause of 3–4 heartbeats between each attempt at worm extraction. The operation was repeated until no worms were visible by echocardiography.

Statistical Analysis

Student t-tests were performed for normally distributed data to identify differences in clinical measurements between the 2 groups. However, the Mann-Whitney U-Test was used for data that were not normally distributed. A P value <.05 was considered significant. The results were expressed as mean ± standard deviation for normally distributed data and as median and range for data that were not normally distributed.


The basket device was used on 22 dogs, of which 18 had caval syndrome (mean age, 7.8 ± 4.2 years; mean body weight, 7.3 ± 4.5 kg) and 4 in class III (mean age, 6.3 ± 4.2 years; mean body, weight 9.1 ± 7.5 kg; Table 1). The mean age and body weight of the animals in this group were 7.3 ± 3.6 years and 8.2 ± 6.3 kg, respectively. The tripod device was used on the remaining 30 dogs, of which 23 had caval syndrome (mean age, 8.8 ± 5.8 years; mean body weight, 9.2 ± 7.4 kg) and 7 were in class III (mean age, 7.8 ± 2.7 years; mean body weight, 12.2 ± 10.3 kg). The mean age and body weight of the animals in this group were 8.3 ± 4.5 years and 10.4 ± 9.8 kg, respectively (Table 1).

Table 1. Summary of the efficacy and complications of heartworm removal in 2 different retrieval devices used in this study. Mean ± standard deviation
  1. a

    Not statistically different mean values or rates. P > .05.

  2. b

    Statistically different mean values or rates. P < .05.

  3. c

    Data not normally distributed.

Dogs (n = 52)22/5230/52
Dogs with class IIIa4/227/30
Dogs with caval syndromea18/2223/30
Age (years)a7.3 ± 3.68.3 ± 4.5
Body weight (kg)a8.2 ± 6.310.4 ± 9.8
Mean worms removed per caseb14.1 ± 7.411.8 ± 5.9
Mean worms removed per attemptb3.5 ± 1.71.9 ± 0.85
Mean attempts to finish heartworm removalb5.5 ± 2.39.7 ± 7.85
Time to complete heartworm removal (minute)b24.5 ± 6.242.3 ± 15.7
Mean number of macerated wormsa,c0.54 ± 0.850.20 ± 0.40
Survival ratea,c21/2224/30
Rate of adverse eventsa,c6/2212/30
Total number of table deatha,c1/223/30
Total number of death after heartworm removala,c0/223/30

The mean number of heartworms removed per case was 14.1 ± 7.4 using the basket and 11.8 ± 5.9 using the tripod, whereas the mean number of worms removed per retrieval attempt was 3.5 ± 1.7 using the basket and 1.9 ± 0.85 using the tripod (Table 1; < .05). The mean number of worms removed was higher in dogs with caval syndrome (13.1 ± 6.5) compared with dogs in class III (7.7 ± 4.1). The mean length of time required to complete a heartworm removal procedure was 24.5 ± 6.2 minutes using the basket, but 42.3 ± 15.7 minutes using the tripod (Table 1; < .05). Furthermore, the mean time to complete heartworm removal was longer in dogs with class III (35.2 ± 7.5 minutes using the basket; 47.8 ± 8.6 minutes using the tripod). In addition, the mean number of attempts to complete heartworm removal was 5.5 ± 2.3 attempts using the basket compared to 9.7 ± 7.85 using the tripod (Table 1; < .05). The mean number of heartworms macerated by a retrieval device was 0.54 ± 0.85 using the basket and 0.20 ± 0.40 using the tripod (Table 1; > .05). The survival rate with the basket was 95.5% (21/22) compared to 80% (24/30) using the tripod (Table 1; > .05). The rate of adverse events (including all intra- and postoperative adverse events such as intraoperative death or postoperative minor bruising) for the basket device was 27.3% (6/22), whereas for the tripod was 40% (12/30) (Table 1; > .05). The adverse events in dogs treated with the basket included chronic coughing from antigen release (2/22) presumed to be in response to heartworm maceration, bruising at the venotomy site (2/22), pulmonary thromboembolism (1/22), and death (1/22). The dog that died was further evaluated and the cause of death was determined to be directly related to hypotensive shock during the procedure. The dog was unstable and in critical condition at presentation, because of heartworm-related acute renal failure and severe pulmonary thromboembolism. Of the 22 dogs treated with the basket, 21 survived to discharge after heartworm removal.

The adverse events in dogs treated with the tripod were chronic coughing presumed to be from antigen release in response to heartworm maceration (2/30), bruising at the venotomy site (2/30), pulmonary thromboembolism (1/30), right atrial tear (1/30), tricuspid valvular damage (1/30), cardiac arrest (2/30), and death (3/30). The causes of death for the 3 dogs were uncontrolled bleeding from a right atrial tear (1/3) and cardiac arrest (2/3). In the 2 dogs that died of cardiac arrest, 1 dog was in critical condition at presentation because of heavy heartworm infection. Of the 30 dogs treated with a tripod grasping forceps, 24 survived to discharge. In addition, 3 dogs that survived the procedure later died in the hospital, despite successful heartworm removal. The causes of death were confirmed by postmortem examination including histopathology and found to result from chronic renal failure because of immune-mediated glomerulonephropathy (1/3), acute respiratory distress from immune-mediated pulmonary disease (1/3), and preexisting pulmonary embolism (1/3).


The retrieval methods described here are introducer-guided heartworm retrieval using 2 types of retrieval devices. Most retrieval devices are difficult to position in the target region, because most lack manipulating aids for placing the tip of the retrieval device and are too rigid to push into the pulmonary vasculature without a preplaced introducer. However, there are some flexible endoscopic grasping forceps that can overcome this obstacle,[10] but most flexible endoscopic grasping forceps (eg, Ishihara alligator forceps) are not applicable and available to small dogs (specifically those <5–7 kg body weight) and more expensive. In the retrieval method described here (ie, introducer-guided heartworm retrieval), any type of retrieval device available on the market can be used, regardless of rigidity, type, and absence of manipulating aids, because these devices can be applied to a preplaced flexible introducer (eg, flexible Tuohy introducer). The preplaced flexible introducer is very flexible and thus easily positioned where the operator wishes. The inner diameter of this introducer is large enough to accommodate most retrieval devices (eg, alligator forceps, basket devices, tripod forceps, snares). Therefore, this introducer-guided retrieval method enables use of most types of retrieval devices and allows efficient control of intraoperative bleeding. Because any disposable retrieval devices (as used in this study) cost approximately USD 150–200 and can be used for this method, the cost of this procedure can be decreased. In fact, most retrieval devices with manipulating aids cost approximately USD 1000–1500. One problem encountered with this procedure is that the worms often are stuck in the introducer when too many worms are caught by the retrieval device, particularly with the basket device. However, this problem can be easily overcome by removing the introducer and the retrieval device simultaneously.

The baskets generally are more flexible and have a wide range of diameters. These devices originally were developed to remove foreign bodies from the esophagus and stomach or stones from the urinary tract.[13] Because baskets do not have sharp grasping claws and cannot grasp the worms, these devices rarely cause intracardiac or vascular damage. Furthermore, baskets can dramatically increase the number of worms captured per retrieval because they work by entrapping the worms instead of grasping them. Our results indicated that the basket was statistically better than the grasping forceps in removing adult heartworms. The mean number of worms removed per retrieval attempt using the basket was twice that using the tripod. In addition, the mean number of attempts for complete heartworm removal for the basket was twice as low as that for the tripod. Therefore, the mean time to complete the heartworm removal with the basket was half that of the tripod. Our study suggests that the efficacy of heartworm removal using the basket is substantially better compared with the tripod. In fact, we noticed that 5~10 worms could be removed in a single attempt with the basket, because the majority of worms within the right cardiac chambers were entangled. Unlike the Ishihara flexible forceps, the basket does not have manipulating aids for placing the tip of the retrieval device. Therefore, heartworms residing in pulmonary arteries cannot be efficiently removed with the basket. We circumvented this problem by developing an introducer-guided removal method as described previously.[10, 12]

The common complications encountered in this study included minor bruising at the venotomy site, chronic coughing from antigen release in response to heartworm maceration, pulmonary thromboembolism, damage to intracardiac structures and cardiac arrest as similarly noted in other studies.[2, 3, 9] We recognize that this method has the added risk of cardiac arrest, presumed to be from cardiac dysfunction caused by the insertion of the introducer in the right ventricle, and especially in dogs <5 kg in body weight. One of the dogs (treated with the tripod) that died of cardiac arrest showed marked cardiac dysfunction (identified fluoroscopically) when the introducer and retrieval device were placed in the right ventricle. Generally, because the grasping type forceps are more rigid, snare or basket-type devices are recommended for small dogs and cats.[11] This finding is corroborated by our study.

Cardiac damage to intracardiac structures (eg, chordae tendinae or right atrium) is an innate problem of heartworm retrieval, because the heartworm is blindly retrieved from the right-sided chamber or pulmonary vasculature.[2, 10] Unfortunately, fluoroscopic guidance can only visualize the retrieval device and not the worms. A recent study reported that transesophageal echocardiography (TEE) could dramatically improve the efficacy of heartworm retrieval and decrease complications.[14] Although we did not use TEE for heartworm removal, we often used routine transthoracic, 2D-echocardiography to visualize live worms and to locate the retrieval device. A left apical 4-chamber view and right parasternal pulmonary artery view, depending on the position of the dog, generally are acceptable for this echocardiographic guidance.

Other complications associated with heartworm maceration by the retrieval device are pulmonary inflammation from rapid antigen release and pulmonary thromboembolism from macerated worms.[15] To decrease these complications, we pretreated all dogs with anti-inflammatory medication (prednisolone) and anticoagulant medication (clopidogrel), although no specific study has yet found those medications helpful to decrease these complications. In addition, doxycycline was administered to prevent possible complications from Wolbachia spp. released from dead worms. Therefore, we believe that such complications were minimized and would not have any life-threatening effect on the animals used in this study. However, a small portion of dogs showed persistent, but minor clinical signs related to these complications, similar to previously reported studies.[2]

Mechanical extraction of adult heartworms generally is recommended in dogs with caval syndrome.[3] Although medical adulticidal treatment generally is recommended for dogs with class III using the split dosage method, mechanical extraction also may be beneficial for these dogs, if the dogs have severe clinical signs associated with HWD. Although melarsomine can effectively kill adult heartworms, the complications (eg, severe respiratory signs, pulmonary thromboembolism) associated with acute death of adult heartworms are anticipated especially in dogs with heavy worm burden, despite premedication (ie, steroids, agents for thrombophylaxis, doxycycline). Therefore, we generally recommend surgery over adulticidal treatment, especially if the dogs had heavy worm burdens with severe clinical signs associated with HWD. In this study, 11/52 dogs were class III; however, they all had severe clinical signs associated with HWD and heavy worm burden in the pulmonary vasculature determined by echocardiography. Furthermore, in some dogs with caval syndrome, the worms sometimes move back into the pulmonary vasculature, although the worms were originally found in the right atrium and ventricle. Therefore, there is enough justification for mechanical extraction for dogs in class III. In our experience, smaller dogs with heavy worm burdens tend to have more severe posttreatment complications associated with adulticidal treatment. Because smaller dog breeds are more popular in Korea, we often treat dogs with interventional methods, despite their not having caval syndrome.

The number of heartworms removed in dogs with class III was less than in dogs with caval syndrome. The reason why the smaller numbers of heartworms were removed in dogs with class III might be because either this class of dogs is less heavily infected with heartworms or more worms may not be removed from the pulmonary vasculature, although we checked postoperatively for complete removal of heartworms by echocardiography. Despite follow-up echocardiography, residual infection still is possible. Therefore, we generally recommended melarsonine treatment for all dogs treated with interventional heartworm removal, after clinical signs of HWD are stabilized.

The mean length of time to complete heartworm removal was longer in dogs with class III because of the difficulty removing heartworms residing in the pulmonary vasculature. Because smaller dogs have a smaller sized cardiac chamber, catheterization of the pulmonary vasculature was technically more difficult and dangerous. Therefore, more careful monitoring of vital signs (eg, ECG for cardiac arrest, arterial pressure for hypotensive crisis) is required for smaller dogs.

In conclusion, our study showed that the basket was more efficient for heartworm removal and that tripod extraction had higher mortality (Table 1). However, our study had insufficient statistical power for a firm conclusion regarding safety. However, we believe these procedures are relatively safe if performed by skilled practitioners with proper retrieval devices in dogs without serious preexisting conditions. Because the operation was performed by 1 surgeon in all cases, there is no bias associated with the skill of the operator. Furthermore, the choice of retrieval device used was made according to the time period (the tripod for the first 3 years and the basket for the remainder of the study period). Therefore, there is no bias associated with the retrieval devices. On the other hand, there may be a minor bias against the basket method in this study because we might have gained more experience using the tripod during the first 3 years, which may have affected the efficiency of worm removal, ultimately narrowing the observed difference between basket and tripod. Because complete removal was only confirmed by echocardiography (not necropsy), the degree of residual infection could not be compared in this study.


The authors thank Prof Clarke Atkins (North Carolina State University) and Dr Lopeti Lavulo (Bioline, Australia) for valuable advice on the manuscript preparation.

Conflict of Interest Declaration: Authors disclose no conflict of interest.


  1. 1

    4DX Heartworm Antigen test, IDEXX, Westbrook, ME

  2. 2

    Daewoo, Seoul, Korea

  3. 3

    Pfizer, Madison, NJ

  4. 4

    Sanofi-Aventis, Cambridge, MA

  5. 5

    Sanofi-Aventis, 500 Kendall Street, Cambridge, MA 02142, USA

  6. 6

    Daewoong, Seoul, Korea

  7. 7

    Daewoo Pharmaceutical, Seoul, Korea

  8. 8

    Daewon Pharmaceutical

  9. 9

    Jeil Pharmaceutical, Seoul, Korea

  10. 10

    Volume-cycled respirator, MDS Matrix 3000, Midmark Corporation, Surrey, UK

  11. 11

    ASafe-T-J Curved fixed core wire guide, 38”×140 cm, Cook Medical, Bloomington, IN

  12. 12

    An introducer sheath, 5–8Fr FlexerTuohy-Borst Side-Arm Introducer; Cook Medical

  13. 13

    Tripod forceps, Rosot Enterprises Inc, Locust Valley, NY

  14. 14

    Disposable fold angular basket, working width 15~30 mm, tube diameter 1.8~2.4 mm, according to the size of dog; Wilson Instruments, Shanghai, P.R. China