Radiographic evaluation of the thorax indicated right heart enlargement. Two-dimensional echocardiography showed right atrial and ventricular dilatation (Figure 1A). The septal leaflet of the tricuspid valve appeared slightly prolapsed. Flattening of the interventricular septum was also observed (Figure 1B). The main pulmonary artery was dilated (main pulmonary artery/aortic diameter = 1.42, reference range <1.0) (9). Echoes coming from filarial worms in the main pulmonary artery were identified (Figure 1C). Enlargement of the tricuspid valve annulus and a concomitant loss of coaptation were noted (Figure 1A). Continuous wave Doppler examination of the tricuspid valve returned a dagger-shaped regurgitant signal characteristic of severe TR (Figures 1D,E). The signal indicates a sharp decrease in the pressure gradient between the right atrium (RA) and RV and consequent elevation of right atrial pressure (10). The distensibility index of the right pulmonary artery was reduced (18.4%, reference range >35%) (11). The acceleration time (AT) was shortened, and the acceleration time/ejection time (AT/ET) ratio was reduced (AT = 50 ms, reference range 52–58 ms; AT/ET = 0.16, reference range >0.3) (1). The caudal vena cava collapse index was as low as 19% (12). A 2.5 L volume of ascites was removed. There was no evidence of respiratory disease or left-to-right shunting causing right ventricular enlargement. In addition, left-sided cardiac disease as a cause of post-capillary PH was not identified. Clinical findings and the clinical history supported a diagnosis of PH and secondary TR along with R-CHF.

Melarsomine is not available in Japan, and moxidectin had already been prescribed for 3 years leading up to initial presentation. Therefore, sildenafil and torsemide were prescribed as treatments for the PH and R-CHF. Despite the volume overload caused by TR and resultant right-sided congestive heart failure, the TAPSE value was low (2.8 mm/kg^0.284) (2). Considering this, we concluded that the right ventricular systolic function was compromised, and we also prescribed pimobendan. Despite increasing the medication dosages [sildenafil (2.4 mg/kg per os, twice daily), pimobendan (0.25 mg/kg per os, twice daily), and torsemide (0.16 mg/kg per os, twice daily)], no improvement in R-CHF was observed. Removal of 2 L of ascites was required approximately every 2 weeks. On the 175th day of follow-up, the dog's activity and appetite decreased, and her general condition deteriorated. Echocardiographic data revealed persistent PH and right-sided volume overload. Except at the initial site, heartworm was not detected using echocardiography, and a heartworm antigen test was negative. In this case, the dog exhibited R-CHF rather than PH symptoms (e.g., syncope). Therefore, we determined volume overload caused by TR to be the main clinical concern. TAP under CPB was recommended. After obtaining the owner's consent, the surgical procedure was performed 221 days after initial presentation.

The anesthesia protocol used was the same as that described in a previous report (13). In preparation for CPB, the right carotid artery was surgically isolated. A right thoracotomy was performed at the 5th intercostal space after administering an intercostal nerve block with bupivacaine hydrochloride. The cranial and caudal vena cava were isolated and encircled with umbilical tape. A purse-string, 5-0 polypropylene suture was placed on the cranial and caudal aspect of the RA, respectively. The azygous vein was isolated and temporarily occluded with umbilical tape. Subsequently, heparin sodium (300 U/kg) was administered intravenously. After 5 min, the activated clotting time was measured and confirmed to be >300 s. An 8-Fr CPB cannula (Flexmate, Toyobo, Shiga, Japan) was inserted into the carotid artery to serve as the arterial line for the CPB. For venous return, a 12-Fr and 14-Fr CPB cannula (DLP Cardiopulmonary Bypass Cannula, Medtronic, Tokyo, Japan) was inserted into the cranial and caudal vena cava via the RA, respectively. CPB was achieved using a heart-lung machine (HAS-II, Senko Medical Instrument, Tokyo, Japan) with an extracorporeal circuit, venous reservoir, and membrane oxygenator (Exelung Kids, Senko Medical Instrument, Tokyo, Japan). After the air was removed from the CPB circuit and the activated clotting time was prolonged to 408 s, a partial CPB was initiated, and the dog's body temperature dropped to 28°C. Blood flow was set at 50–100 mL/kg/min on the CPB pump, and the cranial and caudal vena cava were occluded to initiate total CPB.

Using a right atriotomy approach, intracardiac procedures were performed on the beating heart without administering cardioplegic solution. An incision was made from the RA to the right atrial appendage. The tricuspid valve was viewed through the opened RA (Figures 2A, 3A), and two filarial worms were removed from the RV. There was no elongation of the chordae tendineae in the septum of the tricuspid valve or anterior and posterior leaflets, and no mobility problems were observed (Figures 3A,B). Chordal reconstruction was deemed unnecessary, and only TAP using the De Vega technique was performed.

The De Vega annuloplasty was performed using CV-5 extended polytetrafluoroethylene (e-PTFE) sutures (GORE-TEX Suture, W.L. Gore & Associates, Newark, NJ). One e-PTFE suture with an e-PTFE pledget was started at the anteroseptal commissure and placed around the anterior aspect of the valve (Figures 2B,C). Another identical suture was started over the septal annulus, ~5 mm beyond the posteroseptal commissure, and placed around the lateral aspect of the valve (Figure 2D). Both sutures terminated between the two commissures in the middle of the annulus and were secured using another ePTFE pledget (Figure 2E). Annular reduction was performed by tying down the plication suture while the cylindrical sizer was inserted into the tricuspid valve orifice (Figures 3C,D). The size of the cylindrical sizer was 16 mm, which was set based on the height and width of the septal leaflet (17 and 14 mm, respectively; Figures 2A,E, 3E). A regurgitation test after TAP was performed using saline, and closure of the tricuspid valve was confirmed (Figure 3F). The RA incision was closed using continuous sutures of 6-0 polypropylene (Oval M, Matsudaika Kogyo, Tokyo, Japan). The dog's body temperature was brought back to 37°C, and CPB flow was gradually reduced before being completely turned off. Once the CPB was discontinued and the blood pressure was stable, the catheters were removed from the cranial and caudal vena cava and right carotid artery. A chest tube was inserted through the chest wall, and the cervical and thoracic incisions were closed using standard methods. To antagonize the effects of heparin, 4.5 mg/kg of protamine sulfate was administered intravenously over 30 min. Adequate blood pressure, oxygenation, and ventilation was then confirmed, and the anesthesia was reversed. The mean pre-operative venous pressure of 18 mmHg was decreased to 7 mmHg post-operatively. The durations of extracorporeal circulation, operation, and anesthesia were 70, 187, and 292 min, respectively. The dog recovered well with no anesthesia-related complications, and post-operative echocardiography revealed a reduction in tricuspid regurgitation. The dog was discharged on the eighth post-operative day and prescribed clopidogrel (2 mg/kg per os, once daily for 3 months), rivaroxaban (0.8 mg/kg per os, once daily for 1 week), sildenafil (2.1 mg/kg per os, twice daily), and pimobendan (0.2 mg/kg per os, twice daily).

Follow-up evaluations using echocardiography and chest radiography were performed at 1, 3, and 6 months after surgery.

At the 1-month follow-up, no ascites was observed, and the dog's condition was stable. Echocardiography showed good coaptation of the tricuspid valve and less interventricular septal flattening than was observed during the pre-operative period (Figures 4D,E). The eccentricity index in diastole was decreased (1.34) compared with the measurement taken during the pre-operative period (2.19). The right ventricular end-diastolic volume was reduced from 11.6 cm² before surgery to 7.3 cm². Color Doppler ultrasonography revealed minor tricuspid regurgitation, which was a notable improvement when compared with the pre-operative findings (Figures 4A,B). Compared with the pre-operative recordings, the post-operative fractional area change (FAC) and TAPSE values had decreased from 51.3 to 33.0% and 5.80 to 2.56 mm, respectively. There were no findings suspicious for thrombus formation.

At the 6-month follow-up, the ascites had been completely resolved, and the owner reported that the dog's general condition was good. Echocardiography revealed mild tricuspid regurgitation (Figures 4C,F), but the sizes of the RA and RV had not changed from 1-month follow-up (right ventricular end-diastolic volume = 7.3 cm²). The FAC and TAPSE values were 32% and 2.36 mm, respectively. Although there were improvements in the echocardiographic indices for PH, the presence or absence of PH could not be accurately assessed at follow-up because of reduced RV contractility. Therefore, we continued the treatment for PH beyond the 6-month follow-up. The dose of sildenafil was reduced to 1 mg/kg per os, twice daily, and pimobendan was continued at the same dose (0.2 mg/kg orally twice a day).