[深度剖析] “空蝶鞍”是怎么回事？影像报告中有没有必要提及？会不会因脑脊液搏动而产生的流动伪影呢？...
空蝶鞍（empty sella）一词的定义为蛛网膜下腔延伸至蝶鞍内，因此蝶鞍内并非空无一物，而是填充了脑脊液，其实称其为“鞍内蛛网膜囊肿（intrasellar arachnoidocele）”更为准确，但“空蝶鞍”作为惯用词仍在使用。
空蝶鞍主要的鉴别诊断包括鞍内囊肿和坏死性垂体瘤。空蝶鞍表现为蝶鞍内充满脑脊液，垂体柄延长并与位于鞍底的扁平状垂体相连。垂体柄常被推挤至鞍背前方（图1），在MR矢状位T1加权像上显示较为清楚；而在T2加权像，因受脑脊液流动伪影的干扰而显示不清（图2）。在鞍内囊肿或鞍内肿瘤中，则不会见到这种因脑脊液搏动而产生的流动伪影。有时鞍隔可不完整。影像学检查中常可发现空蝶鞍，随之而来的问题就是：此空蝶鞍为原发性还是继发性？头痛、视力障碍或内分泌功能紊乱是否与之有关？“部分性空蝶鞍”是真正的异常，还是在影像学检查报告中不必提及？需牢记，扁平型垂体，即部分性空蝶鞍，常见于老年人，且激素水平并无异常。在有垂体肿瘤萎缩病史的患者中，无论自发性缩小，还是更常见的药物治疗、卒中、手术或放疗后，诊断继发性空蝶鞍较容易，这些患者的蝶鞍通常扩大。但若不知道既往垂体发生了何种病理过程，则诊断继发性空蝶鞍就较为困难，继发性空蝶鞍可发生于产后垂体坏死、垂体炎或复杂型Rathke囊肿。在这些患者中，蝶鞍形态可能是正常的。大部分继发性空蝶鞍患者中，可见紧贴于鞍底的较小的正常垂体残余，伴或不伴激素水平下降。垂体后叶可异位，或被推挤至鞍背，仅能在轴位T1加权像上观察到。尿崩症罕见，但可见垂体柄阻断效应所致的轻度高催乳素血症（<50ng/ml）。在较大的空蝶鞍中，视神经甚至第三脑室下部均可出现下垂，有时是引起视力症状的原因（图3）。如果患者有头痛，往往与空蝶鞍无明确关联，除非存在慢性良性颅内压增高（图4）。骨质侵蚀可导致瘘口形成和脑脊液鼻漏（图5）。
图1原发性空蝶鞍。
图2继发性空蝶鞍。
图3继发性空蝶鞍，经蝶垂体术后10年。
图4空蝶鞍和良性颅内压增高。
图5 49岁女性，多次分娩史，脑脊液鼻漏、脑膜炎和空蝶鞍。
原发性空蝶鞍与鞍区的任何疾病无关，大部分为偶然发现。但是，其中20%患者，尤其是男性，可出现不同程度的垂体前叶功能低下，特别是生长激素缺乏。鉴于此，对偶然发现的成人（除外老年人）原发性空蝶鞍，建议行内分泌功能评估。原发性或获得性鞍隔缺损非常常见，因此不能解释所有的空蝶鞍。有学者认为，继发于肥胖的慢性脑脊液压力增加，是除了罕见的良性颅内压增高之外的另一个危险因素。多次妊娠在空蝶鞍发生机制中的作用已被肯定，每次妊娠时垂体及蝶鞍都增大，但分娩后垂体回缩，而蝶鞍仍保持轻度扩大状态。最终，相对于垂体来说蝶鞍太大，从而使脑脊液进入鞍内，形成空蝶鞍。我们认为，蝶窦后部气化不全可能是促进空蝶鞍发生的辅助因素，当鞍底与骨松质接触，而不是与蝶窦周边更坚硬的骨皮质接触时，更容易发生蝶鞍扩大。此外，在无症状的蛛网膜下腔普遍扩大，如动眼神经鞘、半月神经节或小脑池等扩大的患者中，也可偶然发现空蝶鞍（图6）。
图6 36岁男性，蛛网膜下腔扩大，原发性空蝶鞍。
The term empty sella refers to the extension ofthe subarachnoid space within the sella turcica;therefore, the sella is not really empty but fi lledwith CSF: the term “intrasellar arachnoidocele”is sometimes preferred, but “empty sella”remains the customary term. Differential diagnosisincludes intrasellar cysts and necrotic pituitaryadenomas. The clue is demonstration of aCSF- fi lled sella with an elongated pituitary stalkattached to a fl attened pituitary gland at the levelof the sellar fl oor. The pituitary stalk is frequentlypressed against the dorsum (Fig.1 )and better seen on coronal T1WI. It can beblurred by fl ow artifacts on T2WI (Fig. 2 ).These fl ow artifacts from CSF pulsations are notseen in cysts or intrasellar tumors. An incompletesellar diaphragm can sometimes be visualized.Demonstration of an empty sella during animaging checkup is frequent and raises questions.Is it a primary or secondary empty sella?Can it be responsible for headache, visual symptoms,or endocrinologic disorders? Is a “partialempty sella” a true abnormality or should it beignored in the radiological report? It is worthremembering here that a fl attened pituitary glandand, therefore, a partial empty sella, is commonin the elderly without any hormonal defi cit.Diagnosis of secondary empty sella is easy incases of a history of pituitary tumor shrinkage,occurring spontaneously or more frequentlyafter medical treatment, apoplexy, surgery, orradiotherapy. In these cases, the sella is usuallyenlarged. Diagnosis of secondary empty sella isless simple if there is no known defi nite pathologicalprocess involving the pituitary gland; itcan be after undiagnosed postpartum pituitarynecrosis, hypophysitis, or complicated Rathkecleft cyst (RCC). In these cases, the sella can beof normal size. In most cases of secondary emptysella, a small normal pituitary gland remnant canbe found at the bottom of the sella, and hormonaldefi cit may or may not be present. The posteriorlobe is either ectopic or pressed against thedorsum sellae, and only visible on axialT1WI. Diabetes insipidus is rare. Mild hyperprolactinemia(<50 ng/ml) resulting from stalkeffect may be noted. In large empty sella, ptosisof optic nerves or even of the inferior part ofthird ventricle may be rarely responsible forunclear visual symptoms (Fig. 3 ). Headache,if present, is usually unrelated to the presence ofempty sella except in the context of benignchronic intracranial hypertension (Fig.4 ).Bone erosion can exceptionally lead to a CSFfi stula and rhinorrhea (Fig. 5 ).
Primary empty sella is unrelated to any pathologyof the sellar region. In most patients, primaryempty sella is an incidental fi nding while in 20 %of patients, mainly men, some degree of anteriorpituitary defi ciency, particularly growth hormonedefi ciency, may be noted. For this reason, an endocrineevaluation can be recommended in adult patients (not in the elderly) on incidental discoveryof a primitive empty sella. Primitive or acquiredsellar diaphragm incompetence is so frequent thatit cannot be held responsible for all empty sellae.Chronic elevated CSF pressure secondary to obesityhas been proposed as a risk factor, in additionto the rare occurrence of benign intracranial hypertension.The role of multiple pregnancies is fairlycertain: at each pregnancy, both the pituitary glandand sella enlarge; the pituitary gland shrinks afterdelivery, but the sella remains slightly enlarged.Ultimately the sella is too large for the pituitarygland, permitting CSF to penetrate the sella. Wehave hypothesized that nonpneumatization of theposterior part of sphenoid sinus could be anadjunctive facilitating factor: deepening of thesella could be facilitated if the sellar fl oor is in contactwith cancellous bone and not with the moreresistant cortical bone bordering the sphenoidsinus. Finally, an empty sella can be found incidentallyin asymptomatic patients with globalenlargement of subarachnoid spaces such as theoculomotor nerve sheaths, the gasserian ganglioncistern, or the cerebellar cistern (Fig.6 ).