A Retrospective Analysis of Venographic Images of a Central Venous Port without Blood Return and Its Usable Period

Abstract

Objective

In routine practice, central venous ports without blood return (CVPWBRs) are common. However, very few studies have reported on the viable period of CVPWBR use. We therefore investigated this period by retrospectively analyzing the venographic images of CVPWBRs.

Methods

We examined patients' venography through the CVPs at the point when they became CVPWBRs for the first time and analyzed the reasons for becoming CVPWBRs. For patients with minor complications of CVPs or normal venographic findings, we used the Kaplan-Meier method to evaluate the period for which such CVPWBRs could be used.

Patients

Eighty-four patients with malignancy whose CVPs became CVPWBRs for the first time between July 31, 2015, and March 12, 2020, were included.

Results

Nine (10.7%) patients had major complications that made the CVPs unusable. Thirty-three (39.3%) patients had minor complications, and the remaining 42 (50.0%) had normal venographic findings. For the 75 patients with minor complications or normal venographic findings who continued to use their CVPWBRs, the Kaplan-Meier method estimated that 25% of complications that might make it unusable would occur within 1,273 days.

Conclusion

There are two learning points in our study. First, venography is needed when the CVP becomes a CVPWBR for the first time due to the high risk, and second, CVPWBRs can be used for a relatively long period in patients without major complications. It is necessary to develop an appropriate follow-up management method for CVPWBRs in prospective studies.

Introduction

Recently, the central venous port (CVP) system has gained popularity not only in the field of oncology but also in other fields within internal medicine. For example, many general physicians use CVPs for home parenteral nutrition in patients with dysphagia, ileus, or terminal diseases. Generally, it is recommended to confirm blood return from the CVP before its use (1).

In daily clinical practice, a CVP without blood return (CVPWBR), in which the blood return cannot be confirmed but infusion through the port is possible, is not a rare phenomenon. To our knowledge, however, there are no reports describing detailed venographic images of CVPWBRs or assessing how long they can be used. In the outpatient chemotherapy center of our hospital, we have a policy of performing venography when a patient's CVP becomes a CVPWBR for the first time.

In the present study, we retrospectively analyzed such venographic images of CVPWBRs to determine the cause of the lack of blood return. In addition, using Kaplan-Meier estimates, we evaluated how long CVPWBRs with normal or slightly abnormal venographic images could be used.

Materials and Methods

Using our hospital medical records, we retrospectively assessed the characteristics of patients with malignancy whose CVPs became CVPWBRs for the first time in our outpatient chemotherapy center. The study was conducted between July 31, 2015, and March 12, 2020. We examined the reasons for CVPWBR and its frequency using the patients' venographic image reports at the time. We classified the status of the CVPs into three types: normal, major complication, and minor complication. Major complications were serious complications that precluded further use of the CVP without prompt medical intervention. Minor complications indicated the presence of some abnormal findings, but the CVP could continue to be used with precaution. CVPs with fibrin sheaths that were still in use after becoming CVPWBRs and were treated with elective stripping or infusion of urokinase were included in the minor complication category.

Furthermore, we selected patients with minor complications or normal venographic findings and evaluated the period for which the CVPWBR could continue to be used. The observation period was from the date of becoming a CVPWBR to May 2, 2021.

This study was approved by the ethics committee of the Osaka International Cancer Institute and complied with the Declaration of Helsinki. Since this was a retrospective study, we provided patients with the opportunity to opt out of the publication of this study.

Statistical analyses

We used descriptive statistics to characterize the patient population (Table 1). Details on the types of complications noted on venographic images when patients' CVPs could not be confirmed to have blood return for the first time are shown in Table 2. The significance of differences between groups, i.e. patients with major complications versus those with minor complications and normal findings, was determined using Fisher's exact test for categorical variables and Wilcoxon's rank sum test for continuous variables. Two-tailed tests of significance were used. In all analyses, p<0.05 was considered statistically significant.

Table 1.

Patient and CV-port Characteristics.

	Number or median or range	%
Sex
Female	39	46.4
Male	45	53.6
Age
Median	64
Range	26-81
Disease
Colorectal cancer	42	50.0
Pancreatic cancer	23	27.4
Breast cancer	10	11.9
Others	9	10.7
CV-port type
Closed end (Groshong)	16	19.3
Open end	58	69.9
Unknown	10	10.8
CV-port site
Left subclavicular vein	13	15.5
Right subclavicular vein	71	84.5
Duration of CV-port use
<6 months	40	47.6
≥6 months-1 year	8	9.5
≥1 year-2 years	21	25.0
≥2 years-3 years	5	6.0
≥3 years	5	6.0
Unknown	5	6.0

CV-port: central venous port

Table 2.

X-ray Images of CV-port without Blood Return.

		Number		%
Major complications		(9)		(10.7)
Displacement of catheter		4		4.8
Tear of catheter		2		2.4
Kink of catheter		1		1.2
Broken port		1		1.2
Large fibrin sheath		1		1.2
Minor complications		(33)		(39.3)
Fibrin sheath		24		28.6
Kink of catheter		2		2.4
Suspected small thrombus		7		8.3
Normal		(42)		(50.0)

Eighty-three cases were diagnosed by venography, and the remaining one case (catheter displacement) was diagnosed by CT scan.

CV-port: central venous port, CT: computed tomography

The Kaplan-Meier method was used to analyze the period for which the CVPWBRs could continue to be used. The cut-off point (event day) of the Kaplan-Meier method was the time when the CVP was considered unusable or unsafe. In this retrospective study, the criteria for these situations included tearing of the catheter, obstruction of the CVP, pain around the CVP, and infection of the CVP (including suspected cases). In the case of tearing of the catheter, the event day was defined as the day when this complication was detected by venography. In the case of the obstruction of the CVP, the day when the obstruction occurred was defined as the event day. The event day of infection, suspected infection, or pain around the CVP was defined as the day when the CVP was removed, as the exact dates of the onset of these events were unclear. The types of events and censored cases in the Kaplan-Meier method were also noted.

All statistical analyses and Kaplan-Meier graphing were performed using the STATA software program, version 16 (Stata, College Station, USA).

Results

Patient and CVP characteristics

A total of 84 patients with malignancy [45 men and 39 women; median age: 64 (range: 26-81) years old] were enrolled during the study period (Table 1). The types of malignancies were as follows: colorectal cancer in 42 patients, pancreatic cancer in 23, breast cancer in 10, and miscellaneous in 9 (non-small-cell lung cancer, 2 cases; gastric cancer, 2; soft tissue sarcoma, 1; duodenal cancer, 1; head and neck cancer, 1; esophageal cancer, 1; neuroendocrine tumor, 1). The types of CVPs were open-end in 58 patients, closed-end (Groshong) in 16, and unknown in 10. CVPs were implanted in the right subclavian veins of 71 patients and the left subclavian veins of 13 patients. The CVP implantation period was <6 months in 40 cases (47.6%), 6 months to <1 year in 8 cases (9.5%), 1 year to <2 years in 21 cases (25.0%), and 2 years to <3 years, ≥3 years, and unknown in 5 cases (6.0%) each. Excluding the 5 cases whose CVP implantation dates were unknown, the median CVP implantation period was 174 (range: 6-1,506) days.

X-ray images of CVPWBRs

Among the 84 patients, 9 (10.7%) had major complications, 33 (39.3%) had minor complications, and the remaining 42 (50.0%) had normal venographic findings (Table 2).

Table 3 shows the details of the patients with major complications. One patient (Patient 7) was judged as having a major complication without having undergone venography because computed tomography performed four days before the onset of CVPWBR revealed catheter displacement. In another patient (Patient 2), the CVP catheter tip moved to the internal jugular vein but then naturally returned to the superior vena cava without any intervention. This CVP was used again without any additional intervention. However, we regard this case as having had a major complication because it disturbed routine chemotherapy.

Table 3.

Patients with Major Complications.

Patient no.		Age		Sex		Diagnosis		Complication		Treatment
1		73		M		Colon cancer		KOC		Replacement
2		65		F		Colon cancer		DOC		No treatment
3		60		M		Colon cancer		Large fibrin sheath		Replacement
4		65		M		Colon cancer		DOC		PRCO
5		58		M		Pancreatic cancer		DOC		PRCO
6		70		M		Colon cancer		TOC		Replacement
7		73		M		Colon cancer		DOC		PRCO
8		66		F		Breast cancer		Broken port		Replacement
9		64		F		Pancreatic cancer		TOC		Replacement

M: male, F: female, KOC: the kink of the catheter, DOC: the displacement of the catheter, TOC: the tear of the catheter, PRCO: position repair by catheter operation

Among major complications, the most common abnormal finding was catheter displacement (4 cases, 4.8% of total CVPWBRs). Furthermore, there were two cases with catheter tears, one with a catheter kink, one with a broken port body, and one with a large fibrin sheath. Six patients had colon cancer; however, there was no significant difference in the cancer type (colon cancer versus other cancers) between patients with major complications and others (minor complications and normal findings) (p=0.48). Similarly, there was no significant difference in the sex or age between the groups.

Among minor complications, the most common finding was fibrin sheath (24 cases, 28.6%). There were also seven cases of suspected small thrombus and two cases of catheter kink in the minor complication group. The seven patients with fibrin sheath subsequently underwent elective fibrin sheath stripping.

In 3 cases in the minor complication category (fibrin sheath), infusion of urokinase (60,000 units) was performed. In the first case, it was used because the CVP became a CVPWBR again following fibrin sheath stripping. In the second case, the CVPWBR with the fibrin sheath continued to be used; however, the drip speed through the CVP eventually decreased, necessitating urokinase use. In the third case, urokinase was used immediately after the first venography session that revealed the fibrin sheath.

Duration of CVPWBR usage

Figure shows the period for which 75 CVPWBRs were able to continue to be used (9 had major complications and could not be used). For the 75 CVPs, the median observation period was 329 days. The Kaplan-Meier method estimated that 25% of events would occur within 1,273 days. There were, in fact, nine events during the observation period (Table 4). Infection was the most common complication, including two confirmed and three suspected cases. In addition, there were two cases of catheter tears, one case of obstruction, and one case of pain around the catheter.

Figure.

A Kaplan-Meier analysis of the duration of CVPWBRs. The percentage of usable CVPs from the day they first became CVPWBRs is shown. CVPs: central venous ports, CVPWBRs: central venous ports without blood return

Table 4.

Types of Events in the Kaplan-Meier Method.

		Number		%
Confirmed infection		2		22.2
Suspicion of infection		3		33.3
Obstruction		1		11.1
Pain around CV-port		1		11.1
Tear of catheter		2		22.2

CV-port: central venous port

Table 5 shows the types of sensors in the Kaplan-Meier method. The most common type was the end of patients' follow-up (39 cases, 59.1%). The two most common reasons for discontinuing the follow-up were transfer to another hospital or clinic (34 cases, 51.5%) and continued use of the CVP until the end of the observation period of this study (5 cases, 7.6%).

Table 5.

Types of Sensors of the Kaplan-Meier Method.

		Number		%
Discontinuation of follow-up		39		59.1
Patient death		6		9.1
Catheters removed when no longer needed		6		9.1
Maintenance cessation for unknown reason		15		22.7

Discussion

A fibrin sheath is considered to be one of the causes of CVPWBRs (2), but the exact mechanism underlying CVPWBR development remains unknown. More worrisome is the fact that some serious CVPWBR complications have been reported (3); we therefore cannot ignore adverse events like CVPWBR development.

We believe that there are two learning points in our study. In the present study, 9 cases (10.7%) had major complications; thus, the first learning point is that conducting venography through the CVP at the point that it becomes a CVPWBR for the first time is indispensable. There is no notable trend in the occurrence of major complications and characteristics of patients, such as disease, age, and sex; therefore, it is necessary to pay attention to all patients. In contrast, among patients with minor complications or normal venographic findings, 75% were able to continue to use their CVPWBR for more than 3 years. We believe that this is a relatively satisfactory period in the field of oncology; this is the second learning point of our study.

A previous report on the reasons for the removal of CVPs showed that among the 71 patients whose CVPs were implanted for various purposes and removed, 39 were removed due to catheter-related bloodstream infection (CRBSI), including 26 suspected cases of CRBSI (4). This report also showed that there were 11 cases where CVPs were removed due to complications other than CRBSI. In our study, among the 75 CVPWBRs that were still being used, 5 were removed due to infection, including 3 suspected cases, making infection the most common cause of removal. Thus, infection management should be given the utmost priority when using CVPWBRs. Many upper-extremity deep-vein thromboses are associated with CV catheters (5). In our study, there were no obvious cases of upper-extremity deep-vein thrombosis, possibly due to the elective fibrin sheath stripping performed in 7 out of 24 cases with fibrin sheaths.

The need for treatment of CVPWBR is unclear. As a treatment for catheter obstruction, the replacement or the use of thrombolytic drugs such as reteplase, alteplase, or urokinase is suggested (6-8). The replacement of a CVP is a highly invasive procedure, especially for patients with malignancy. Furthermore, reteplase or alteplase is generally expensive, and the health insurance system of the Japanese Ministry of Health, Labour and Welfare does not cover these drugs when used for the recovery of CVPs. Urokinase is also not approved for this use in the Japanese insurance system. However, the efficiency of urokinase for the treatment of fibrinous obstruction of venous access devices has been reported (9), and it is inexpensive compared with other thrombolytic drugs. Therefore, since September 2018, urokinase has been used at our institution when a fibrin sheath is detected on venography.

No appropriate follow-up management for CVPWBR has yet been established. Venography through the CVP is the gold standard of diagnostic imaging; however, it is invasive and cannot be repeated in daily clinical practice. It has been reported that, as a diagnostic test for upper-extremity deep-vein thrombosis, compression ultrasonography may be a viable alternative to venography (10). However, to our knowledge, there are no reports of ultrasonography for CVPWBR. In the future, more research on ultrasound images of CVPWBR may be necessary.

In our study, as follow-up management for CVPWBRs that were still in use, nursing staff recorded the drip rate and resistance on flushing the CVP in the medical record. Any drastic change in these parameters from the prior recorded value necessitated another venographic examination. This management method is quite subjective, which is a limitation of our study. In many malignant diseases, abnormalities in coagulation are common phenomena, which might have affected the occurrence of CVP complications. However, we cannot evaluate this relationship because most patients did not undergo regular hemostatic tests in our retrospective study, which is another limitation of our study.

In summary, venography through the CVP at the time it first becomes a CVPWBR is required because of the high rate of serious complications. However, in cases where venographic images show a minor complication or normal findings, the CVPWBR can be used for a relatively long time. It is necessary to develop an appropriate, non-invasive follow-up management method for CVPWBRs in prospective studies.

The authors state that they have no Conflict of Interest (COI).