The Utility of the DDAVP challenge test in children with low von Willebrand factor (VWF)

Low von Willebrand factor (VWF), defined as either VWF antigen (VWF:Ag) or Ristocetin cofactor (VWF:RCo) level ≥ 30 and < 50 iu/dl, is a common finding in paediatric patients tested for von Willebrand Disease (VWD), the most common inherited bleeding disorder. DDAVP (1-deamino-8-D-arginine vasopressin, desmopressin), a synthetic derivative of vasopressin that promotes the release of VWF multimers from Weibel-Palade bodies in the vascular endothelium, is safe and effective in preventing and treating bleeding in children with VWF levels < 50 iu/dl (Gilly, et al 2002, Khair, et al 2007, Leissinger, et al 2001). In our paediatric population, children are often tested for VWD due to a family history of VWD, personal bleeding history or prolonged partial thromboplastin time, as well as, but not limited to, part of a pre-operative evaluation. If found to have VWF:Ag or VWF:RCo <50 iu/dl, patients automatically undergo a DDAVP challenge test to assess laboratory response prior to the administration of DDAVP for the prevention or treatment of bleeding.

In our paediatric hematology department, a total of 204 children underwent DDAVP challenge testing between 1 January,2005 and 31 December, 2013. We defined complete response to the DDAVP challenge test, as a two-fold increase in both VWF:RCo and factor VIII (FVIII) coagulant (FVIII:C) or postresponse levels > 100 iu/dl, based on previous literature and targeted post-response levels of North American physicians (Cohen, et al 2001).

Of the 204 total children tested, the majority (162, 79·4%) had low VWF while 42 (20·6%) had type 1 VWD. One hundred and ninety patients (93·1%, 95% confidence interval: 89·6% to 96·6%) had a complete response to DDAVP. All non-responders failed the DDAVP challenge based on circulating post-response levels of VWF:RCo (Table I). The desmopressin route of administration was known in 184 children, 94% of whom received DDAVP intravenously. There were no differences in gender, age, race, ethnicity or blood groups between the children with low VWF and type 1 VWD.

Table I.

Characteristics of non-responders with low von Willebrand factor and type 1 von Willebrand disease

Group	Age (years)	Sex	DDAVP	Race	Ethnicity	Blood Group	VWF:RCo		VWF:Ag		FVIII:C
							pre	post	pre	post	pre	post
LVWF	13	M	IV	White	NH	A	39	55	76	144	86	69
	5	M	IV	White	NH	A	58*	75*	95	308	105	63
	15	M	IV	Unknown	Unknown	O	50*	63*	48	84	84	75
	4	M	IV	Unknown	Unknown	O	44*	55*	67	98	69	66
	14	F	Nasal	White	NH	O	31	35	77	117	68	61
	17	F	Nasal	Black	NH	O	41	56	63	99	81	59
	15	F	Nasal	Unknown	Unknown	O	38	51	78	79	50	42
	4	F	Unknown	White	NH	O	39	50	60	228	94	77
Type 1 VWD	4	M	IV	Asian	NH	O	33†	39†	81	91	45	41
	3	M	IV	Unknown	NH	AB	35†	47	105	123	55	40
	3	M	IV	White	NH	O	20	13	27	48	25	21
	4	M	IV	Asian	NH	O	20	14	50	88	29	35
	7	F	Nasal	White	Unknown	O	39†	67	70	95	82	54
	13	F	Unknown	White	Unknown	O	25	35	70	118	80	48

LVWF: low von Willebrand factor; VWD, von Willebrand disease; NH: non-Hispanic

^*prior value in our electronic medical record between 30-50 iu/dl

^†prior value in our electronic medical record less than 30 iu/dl

Looking specifically at those with low VWF, 154 (95%) children had a complete response. Figure 1 shows the VWF:Ag, VWF:RCo and FVIII:C levels for children with low VWF tested at baseline and at 1 h after the end of infusion.

Figure 1. Patient Pre- and Post- VWF:RCo, VWF:Ag and FVIII:C levels.

Pre- and post-VWF:RCo (left), VWF:Ag (middle) and FVIII:C (right) values for all 162 patients tested at baseline and at 1 h after DDAVP infusion. All non-responders (open circle) failed the DDAVP challenge based on post-response levels of VWF:RCo.

VWF:RCo, Ristocetin cofactor; VWF:Ag, Von Willebrand factor antigen; FVIII:C, factor VIII coagulant; DDAVP, desmopressin [1-deamino-8-D-arginine vasopressin]).

The DDAVP response rate seen in our study is similar to that reported in previous studies, despite the use of different definitions for low VWF, type 1 VWD and DDAVP response criteria in these studies.(Castaman, et al 2008, Revel-Vilk, et al 2003, Sánchez-Luceros, et al 2010) Sanchez-Luceros et al (2010) used a less conservative response criteria of a two-fold increase in post-VWF:RCo and FVIII:C or post values > 50 iu/dl to define DDAVP response . Using these criteria, 99·4% of the children reported in this study had a complete response to DDAVP. Importantly, 7 out of 8 (88%) of the low VWF non-responders would have been classified as responders. The one remaining low VWF non-responder used intranasal DDAVP for the challenge test. Most children will experience facial flushing with intranasal DDAVP (Khair, et al 2007). In this patient's hospital record, it is noted that the patient had no facial flushing and that the providers did not feel the patient received a full dose. Thus, if the criteria of Sanchez-Luceros et al (2010) were applied to this study population, all of the children with low VWF would have had a complete response after receiving an appropriate dose of DDAVP.

The monetary savings associated with eliminating the DDAVP challenge in this subset of individuals is one of several potential benefits. Patients often receive FVIII concentrates for bleeding if a DDAVP challenge has not yet been performed. The use of DDAVP without an obligatory DDAVP challenge test in patients with low VWF decreases this population's potential exposure to plasma-derived products, which are not only costly but carry hypothetical infectious risks. The additional required blood draws, intravenous line placement and intravenous infusions for the DDAVP challenge test are also anxiety provoking, painful experiences in children. Minimizing intravenous DDAVP therapy to only episodes of bleeding or bleeding prevention will decrease the risk of potential side effects, including headache, emesis and hyponatraemia as well as the emotional stress associated with the diagnosis and care of paediatric patients with low VWF.

The most recent United Kingdom Haemophilia Centre Doctors Organization guidelines for the diagnosis and management of VWD made no clear recommendations regarding patients with low VWF (Laffan, et al 2014). While our study is limited by its retrospective design, small patient population and lack of subsequent post-DDAVP challenge results beyond 1-h, it clearly shows that given the high DDAVP response rate in children with low VWF, the use of the DDAVP challenge tests among paediatric patients with VWF levels ≥ 30 iu/dl and < 50 iu/dl should be reduced, if not eliminated. While our study was not designed to determine which components of the DDAVP challenge test are most useful, our data suggests that consideration could be made for only post-DDAVP VWF:RCo levels to assess response in patients who clinicians deem require a DDAVP challenge. Limiting the DDAVP challenge laboratory response panel to post-VWF:RCo would result in a significant cost reduction. An important question not addressed here, however, is whether reducing or eliminating the DDAVP challenge may result in increased bleeding and associated health effects and treatment costs. Other studies have reported a very low risk of bleeding and lack of correlation between risk of bleeding and DDAVP challenge response ,which would suggest that current practice to perform the DDAVP challenge test among this group of patients may not be a good use of limited health care resources (Castaman, et al 1999, Rodriguez, et al 2010). Nevertheless, further prospective studies should evaluate that trade-off between savings and increased risks, if any, within this group.