Removal of the needles after a hemodialysis session in patients with arteriovenous fistulas (AVF) is as important as the cannulation process. An adequate technique protects the wall of fistula from possible injuries and favors adequate hemostasis.1–3

The application of correct pression on the puncture points favors their closure and reduces the risk of complications, such as hematomas, stenosis or thrombosis.1,4–6 These adverse effects makes subsequent punctures difficult, limits the number of options1,4 and, finally, the management of these patients becomes difficult.

For the process of hemostasis, the main clinical guidelines recommend applying compression with the fingertips for at least 10 min after removal of the needles, the pressure has to be applied constantly and maintained until complete hemostasia is achieved.7–9

The use of compression devices (clamps, bands) is also accepted in some of these guidelines,2,7 while others prohibit or discourage their use,1,4,8 arguing that these devices could apply excessive pressure and generate damage of the fistula. This recommendation is based on expert consensus, but to our knowledge, there are no studies that provide data on which method of hemostasis is more appropriate.

Despite the negative consideration of the use of compression devices in AVF (especially in prosthetics), these are used in clinical practice. There is a large number of this type devices available on the market. Many of them have a regulation system that allows adjusting the compression exerted that could make them suitable for use in AVFs.

It would be useful to have data on the feasibility and safety of the use of this type of device in clinical practice, as well as its comparison with manual compression. The objectives of the present study are: (a) to assess the compression exerted with an adjustable clamp and compare it with that exerted manually, and (b) to assess the intensity and variations of the pressure that is exerted manually on the AVF during the hemostasis process.

Descriptive cross-sectional study comparing two compression methods in patients undergoing hemodialysis through native or prosthetic AVFs in the Hemodialysis Service of the Segovia Health Care Complex.

Data was collected from a total of 51 hemodialysis sessions in15 patients (mean age, 71.33 ± 12.66 years, 9 females). The mean value of urea prior to the hemodialysis session was 113.41 mg/dl (range 88−180 mg/dl). The average Kt was 48.42 ± 5.70 l (range 36−60 l). A value of Kt below the optimum was observed in 2 (3.9%) sessions. The mean value of hemoglobin before hemodialysis was 10.81 ± 1.21 g/dl (range 9.60–13.20 g/dl).

There were 4 patients that did not complete the study: one was discontinued for presenting critical stenosis in the venous anastomosis and 3 patients died (one sudden death and 2 due to advanced neoplasia). The reasons for withdrawal of these 4 patients were not associated with the hemostasis process in the AVF.

Out of the 15 patients included, 2 (13.3%) had a Gore-Tex prosthetic fistula implanted, both of humerus-axillary location, and results were obtained from 9 (17.6%) of the 51 sessions evaluated. The remaining 13 (86.6%) patients had a native fistula, in which data from 42 (82.3%) sessions were collected. Of these 13 fistulas, 8 were humerus-cephalic, 3 radio-cephalic, and 2 humerus-basilic. The median age (range) of the fistulas was 592 (3,589) days.

Regarding the usual daily treatment, 6 patients were on of acetylsalicylic acid 100 mg, 2 on 100 mg of acetylsalicylic acid and 75 mg of clopidogrel, and 2 were anticoagulated with 20 mg of enoxaparin. All patients were anticoagulated with sodium enoxaparin at the beginning of the hemodialysis session, with doses between 20 and 80 mg. All patients received erythropoietin as a treatment for anemia. During the development of the study, none required transfusion of packed red blood cells. In all the sessions analyzed, an absorbable gelatin dressing (Surgispon®) was used as coadjuvant hemostatic agent to apply pressure to the puncture site, both with the forceps and with the patient’s fingers.

Of the 79 hemodialysis sessions initially included in the study, 2 (2.5%) were excluded due to the fibrin deposition in the system that prevented the correct reading of intra-access pressure and 26 (32.9%) due to movements or changes in the position of the patient that affected the reliability of the readings. Therefore, in the final analysis, the data from 51 sessions were considered. In none of them there were complications that could be associated with the hemostasia process.

The mean values ​​and standard deviations of the intra-access pressure in each of the readings are shown in Table 1, and the mean values ​​of the differences between each of the readings are presented in Table 2, The distribution of these values ​​are shown in Figs. 1 and 2.

Fig. 1.

Distribution of values intra-access pressure ​​(mmHg).

Basal: without exerting compression on the arteriovenous fistula; M0: at the start of manual compression; M3: minute 3 of manual compression; M6: minute 6 of manual compression; Q1: when placing the clamp, without adjusting it; P2: after adjusting the compression using the adjustable system.

(0.05MB).

Fig. 2.

Distribution of the values ​​of differences of the intra-access pressure (mmHg).

Basal: without exerting compression on the arteriovenous fistula; M0: at the start of manual compression; M3: minute 3 of manual compression; M6: minute 6 of manual compression; Q1: when placing the clamp, without being adjusted; P2: after adjusting the compression using the adjustable system.

The differences between basal and manual pressure are shown in white. Medium gray shows the differences between basal pressure and adjustable clamp. Differences between adjustable clamp and manual pressure are shown in dark gray.

(0.06MB).

When placing the clamp to perform hemostasis, and before being adjusted by the nurse (P1), the increase in pressure was +31.31 mmHg (+116.56 %, p < 0.001) with respect to the basal pressure. Subsequently, after being adjusted by the nurse (P2), the pressure decreased −16.82 mmHg (−28.91% compared to P1, p < 0.001). At that time (P2) represented an increase of +14.49 mmHg (+53.94%, p < 0.001) with respect to the basal pressure.

The increase in pressure when puncture point was first pressed manually (M0) was +23.92 mmHg (p < 0.001), representing an increase of +89.05% as compared to baseline intra-access pressure. From that moment, the pressure exerted showed a downward trend, with an average decrease of −8.82 mmHg (−17.36%, p < 0.001) at 3 min (M3) and −12, 55 mmHg (−24.71%, p < 0.001) at 6 min (M6).

After the clamp was adjusted by the nurse the pressure (P2) was lower than the values registered with the patient's manual pressure at minute zero (M0); the mean difference was -9.43 mmHg (−18.57%, p < 0.001). However, it was observed that these values tended to be equal with M3 (−0.61 mmHg, −1.45%, p = 0.901) and M6 (+3,12 mmHg, +8.1%, p = 0.102), with a very similar distribution of values.

None of the patients included in the study showed complications associated with the process of hemostasis within 48 h after the puncture of the fistula.

The results of this study show that the pressure exerted on the AVF by an adjustable compression system is equal to or less than that exerted manually by the patient. In addition, it has been shown that the manual pressure applied by the patient is not uniform over time and shows a descending pattern as the hemostasis process progresses.

Although these results may seem somewhat predictable, they needed to be evaluated since there is not objective data in the literature and, there is not information about which method of hemostasis would be more convenient. In fact, due to this lack of data, the recommendations by clinical guidelines are currently based only on the consensus of experts.

The recommendations assume that manual compression is the most appropriate and safe, since it is moderate and constant over time.2,12 However, our data reveal that manual pressure is not constant but decreases over time, possibly due to a reduction of tone or to circumstances that decrease the concentration of the patient. One of the advantages of compression devices would be that they would not be affected by these factors.

The use of compression devices in AVF has a negative consideration in the guidelines, so it is generally discouraged, or allowed with restrictions in some situations.1,2,8,13 This is based on the premise that they may exert excessive pressure on the fistula, favor the development of complications and shorten its survival. In any case, as our data suggests, this appears not to be the case with all devices. The risk of a device applying inadequate compression to achieve hemostasis (by excess or by defect) is only expected in those devices that do not allow pressure adjustment, or in adjustable devices that are not adequately supervised.

As our results show, the pressure exerted by a device that allows adjustment is similar to or less than that exerted manually by the patient. This reveals several issues: (1) that the premise of excessive pressure, assumed in clinical guidelines, would not be true in all devices, only in those that do not have a system that allows adjustment; (2) the importance of being used by trained personnel, capable of determining the appropriate compression in each patient and in each session, and (3) the need of a system that allows adjustment compression in all devices marketed for this purpose.

However, despite the usefulness of the data provided in this study, they do not allow us to know the effect of the regular use of these devices in aspects as relevant as the occurrence of complications and survival of the fistula. These are issues that must be addressed in future clinical trials, the results of which would provide the first evidence on which to base the recommendations of the clinical guidelines.

Although our results do not allow us to provide data regarding the safety of the continued use of these devices, we did observe an absence of complications in the hemostasis process. This has been the case even in patients with a more unfavorable uremic situation that has been associated with a greater risk of bleeding.14

Finally, it should be mentioned that, although the compression method is an important factor in determining the success of the hemostasis process, it also depends on a correct cannulation technique which includes, the use of the most appropriate type of needle for each patient (length, gauge, fenestration),15 the use of the most appropriate technique16 or the performance of an ultrasound-guided puncture when necessary.17 This can reduce the risk of a troublesome hemostasis process and the appearance of complications.

Our study has some limitations. Although we have used a sample size sufficient to determine the existence of pressure differences between the two methods, this study has not evaluated the safety of the regular use of adjustable devices and its effects on the lifespan of the fistulas, an issue that has been commented already. Furthermore, the small number of prosthetic fistulas included in the study did not allow us to analyze whether the results obtained would be applicable to this type of fistula. In addition, we have evaluated only one of the many compression devices on the market. Although we think that our conclusions could be extended to any adjustable device, there could be differences depending on the type of regulation system used (continuous band,18 discontinuous band,19 bandage20 or clamp21 ). We have evaluated a clamp system, that could have disadvantages as compared to other systems, such as the possibility of displacement during the hemostasis process. In this regard, it would be interesting to be able to analyze the use of compression band systems, which could be more effective in avoiding this problem.

To conclude, the results of this study have shown that the intensity of manual compression on the AVF decreases throughout the hemostasis process, and that the compression exerted by a forceps after being adjusted by the nurse is lower or similar to that applied manually. These data suggest that the premises on which some of the recommendations in the clinical guidelines are based could be imprecise, such as that the compression exerted manually is constant over time, or that all the devices exert excessive compression on the fistula. However, the role of routine use of adjustable compression devices in patient safety and fistula survival are issues that still need to be evaluated in future clinical trials.

None.