Chapter 9: Bilaterality

---

9.1 Imaging Problems

Trish Chudleigh⁴⁹ noted after five cases of CPC in normal fetuses that they are “probably always bilateral”, though only two patients in her series were “obviously bilateral”. She noted that “reverberation within the fetal skull means that they are always visualised with greater ease in the distal [i.e. furthest from the transducer] cerebral hemisphere.” Ostlere¹⁴⁶ noted this tendency; 75% of solitary cysts in his series were in the lower hemisphere. Several authors, when discussing bilaterality felt that in a significant number of their cases no comment could be made as the upper hemisphere was not well seen. Chinn⁴⁶, for example, could make no comment in 19 of his 38 cases. DeRoo⁶² was unable to show the upper hemisphere in 6/12 fetuses. As with cyst incidence, there may be some relation between the ability to see the upper hemisphere and the quality of the ultrasound machine being used, as well as the size of the patient (fat patients produce more reverberation artefact and the beam is defocused), and the skill of the sonographer or sonologist performing the scan.

This remains one of the most popular criteria, and many sonographers and sonologists I have spoken to seem to consider it significant. Despite Chudleigh’s initial thought that when present, cysts were invariably bilateral, many researchers have had difficulty finding cysts in the upper hemisphere, with the result that only 54% have been reported as bilateral cysts. There is considerable discrepancy of the rates of bilaterality amongst reports however, and the extremes are mentioned as follows:

Author (Brand of Equipment)	Incidence of Bilaterality
Thorpe-Beeston^185 (?Aloka)	82/83 (98%)
Achiron^2 (Elscint)	27/30 (90%)
Chinn^46 (Acuson)	8/38 (21%) Uncertain in 19
Platt^152 (Acuson)	13/71 (18%)
Geelong Hospital (Aloka, ATL, Diasonics)	28/50 (56%)
All Others:	~ 50%

Table 10: Incidence of bilaterality is quite varied.

9.2 Lower Hemisphere Preponderance

Of these unilateral cysts in normal fetuses, the majority have been in the lower hemisphere. As mentioned, several investigators were unable to review the upper hemisphere adequately. In the Geelong Hospital series, 15/22 or 67% of unilateral cysts were in the lower hemisphere. With our protocol of defining a cyst in two planes whenever possible, many unsuspected extra cysts were noted in the upper hemisphere, particularly when coronal imaging was employed. The usual artefacts are slightly less in this plane of section, and this is also true when an infero-superior angle of approach is achieved through the line of the lambdoid suture. Similarly, coronal imaging from the superior aspect through the anterior fontanelle or from behind through the posterior fontanelle and the skull sutures provides excellent demonstration of the lateral ventricles in many instances.

The implication of these scanning techniques is that the upper hemisphere has not been assessed as well as the lower hemisphere: there should be a 50% split of upper and lower cysts if they were truly unilateral. It is therefore possible that many of the cysts that are called unilateral are in fact bilateral. It has been clearly demonstrated that certainly not all CPC are bilateral.

9.3 Statistical Analysis

Of all CPC reported, 1,229 have mention made of this criterion. In normal fetuses, 551 were bilateral. In aneuploid fetuses, 60 were bilateral and 40 were unilateral. Analysis of these figures shows that bilaterality has a sensitivity of 60% and a specificity of 51.2% (equivalent to flipping a coin). This means that nearly half of patients with aneuploidy would be missed, and half of the patients with normal fetuses would show a false positive result.

The authors in whose articles it was possible to determine the bilaterality of normal and aneuploid fetuses are shown below.

Author	No. Cysts	Bilateral Euploid	No. Aneuploid (T18)	Bilateral Aneuploid (T18)
Nicolaides^131	4		1	1
Bundy^33	1		1	1
Furness^81	30		3	1
Chitkara^47	41	20	11	1
Clark^50	5	2
DeRoo^62	17	4
Benacerraf^14	38	16
Gabrielli^83	82	44	4	1
Chan^45	13	4
Khouzam^111	1		1	1
Camurri^40	10	1	1	1
Ostlere^146	100	52	3	3
Thorpe-Beeston^185	83	62	20	19
Chinn^46	38	8	1
Twining^189	19	8	2
Platt^152	71	13	4
Achiron^2	30		5	4
Rotmensch^163	1		1	1
Perpignano^150	87		6	4
Nadel^123	234	125	12	6
Porto^154	63	28	6	4
Kennedy^110	22	9	3	2
Burrows^34	1		1	1
Oettinger	14		2	1
Nava	211	85	8	5
Gross^91	80	45	2
Kupfermine^116	102	25	7	1
Walkinshaw^193	152		4	2
Total (T18)	1,229	551	100 (73)	60 (52)

Table 11: Authors who describe CPC according to bilaterality.

Fig 13: Bilaterality in T18 and Normal fetuses.

	T18	Normal	PPV
Bilateral	52	551	8.6%
Unilateral	21	578	Likelihood ratio
Total	73	1229	0.0146 (1.46%)
Sensitivity	71.2%	51.2%	Specificity
False Negative Rate	28.8%	48.8%	False Positive Rate

Table 12: Bilaterality: 2 x 2 table.

The sensitivity and specificity are very low with this criterion, and it is not surprising that, using the chi-squared test, the difference in bilaterality shows only mild statistical significance (p = 0.032) between euploid and aneuploid fetuses. The odds ratio is calculated to be 2.59. The difference in odds is calculated at 0.2 (CI -0.09 to 0.31). Because of the negative odds difference at the 5% confidence interval (an impossibility), there is a possibility, therefore, that these results are all due to chance.

Again using Bayes theorem to include the prevalence of trisomy 18, the equation becomes:

Odds (Risk) of T18 in bilateral cysts = 

0.0433% x 1.46% = 0.065% or 1 in 1542.

From this calculation it can be seen that despite the mild statistical significance of bilaterality, the risk for T18 in bilateral cysts is only slightly greater than the risk for T18 alone. Further breakdown of this risk by maternal age is provided in the Appendix.

9.4 Number of Cysts – Confusion of Terminology

Closely allied with the criterion of bilaterality is the number of cysts seen. Only four aneuploid fetuses have been categorically noted to have solitary cysts but this figure cannot be relied on as many cysts described as unilateral cysts are probably, but not necessarily, solitary, and there may be missed cysts in the upper hemisphere. In most reports cysts are not described in terms of number. Furness⁸¹ referred to a case where cysts “completely replaced the choroid plexus bilaterally”, in a case of T18, the accompanying images of which showed multiple cysts of various sizes.

Confusion easily clouds the discussion. While unilateral cysts also can be multiple, solitary or single, one wonders how bilateral cysts may be solitary or single. They can all be isolated. Consistent definitions are not adhered to: Achiron refers to “bilateral solitary” cysts. In my reading the following terms have been used and the meaning that usually applies from the context is as given (Table 13):

unilateral cysts:	single or multiple cysts seen on one side only
bilateral cysts:	single or multiple cysts on both sides
multiple cysts:	more than one cyst (on one side)
solitary cyst:	1. one cyst 2. a single cyst on one side in cases of bilateral cysts 3. cyst which is the only detected anomaly
single cyst:	one cyst
isolated cysts:	cysts which are the only detected anomaly

Table 13: Types of numericity of cysts.

In order to avoid confusion, we feel that discussion over the number of cysts should be avoided. The low numbers of aneuploid fetuses with a single cyst and the uncertainty of definition make it impossible to analyze this criterion meaningfully. Along with bilaterality it suffers from the problem of upper hemisphere visualization and therefore the numbers that are given are undoubtedly wrong.

9.5 Conclusion:

The use of bilaterality as a criterion suffers for the following reasons:

· technique and equipment greatly affect visualization of upper hemisphere therefore its true incidence cannot be established.

· low sensitivity and poor specificity.

· difference between normal and aneuploid fetuses shows only weak statistically significant, and may be due to chance.

· adjustment of risk by Bayes Theorem shows the risk with bilateral cysts is very similar to the initial risk of T18, and is not sufficient to indicate amniocentesis in the median maternal age range.

The ability to discriminate aneuploidy on this criterion is obviously severely limited by technical problems of demonstrating the upper lateral ventricle, and the statistics analysis shows that it is not a practically applicable criterion in any given case.

---